CN119915352A - Inspection data monitoring method and server for early warning of roads and bridges in mountainous areas - Google Patents

Abstract

The present invention provides a patrol data monitoring method and server for early warning of roads and bridges in mountainous areas. First, remote sensing monitoring data of roads and bridges including current terrain distribution characteristics, road and bridge structural state parameters and historical disaster records are obtained. Then, according to the current terrain distribution characteristics, a terrain feature extraction model is called to segment the target mountainous road section, and patrol sub-areas and terrain feature labels are generated. Then, according to the terrain feature labels and the road and bridge structural state parameters, a dynamic flight path including patrol nodes and flight parameters is generated for an unmanned aerial vehicle through a path planning model. Subsequently, the unmanned aerial vehicle is controlled to patrol along the path, and multi-dimensional monitoring data such as surface deformation data, road and bridge crack images and slope displacement trajectories are collected. Finally, the multi-dimensional monitoring data and historical disaster records are input into a risk early warning model, and risk level signals and disposal suggestion information are generated, so as to realize comprehensive intelligent monitoring and early warning of roads and bridges in mountainous areas and improve monitoring accuracy and efficiency.

Description

Patrol data monitoring method and server applied to mountain road and bridge early warning

Technical Field

The invention relates to the technical field of computers, in particular to a patrol data monitoring method and a server applied to mountain road and bridge early warning.

Background

In the mountain road and bridge construction and operation process, ensuring the safety and stability of the road and bridge is important for ensuring smooth transportation and life and property safety of people. Because the mountain area geographical environment is complicated, the topography is fluctuant big, the geological condition is various, and receive natural disasters such as landslide, mud-rock flow, earthquake etc. often to influence, mountain area road bridge face is a great deal of potential risk, therefore carries out effectual monitoring early warning to mountain area road bridge and is especially crucial.

Currently, there are various ways to monitor mountain roads and bridges. Traditional manual inspection mode relies on inspection personnel to view road and bridge conditions in the field. However, the mountain area has complex topography, the manual inspection is not only inefficient, but also difficult to reach some areas with dangerous features, inspection blind areas are easy to appear, and the overall condition of the road and bridge cannot be comprehensively and accurately mastered.

As technology evolves, some fixed sensor based monitoring systems are applied. These monitoring systems acquire partial data by installing fixed sensors at strategic locations of the road and bridge, such as simple displacement sensors to monitor changes in the displacement of the road and bridge. However, the method has obvious limitation, the mounting position of the sensor is fixed, the monitoring range and the key point can not be flexibly adjusted according to the actual risk condition and the terrain characteristics of the road and the bridge, and the risk monitoring capability for the complex terrain environment around the road and the bridge and the dynamic change of different areas is insufficient.

Meanwhile, in the aspect of routing inspection path planning, most of the prior art adopts a mode of presetting a fixed route, whether manual routing inspection or routing inspection by using unmanned aerial vehicle and other equipment, the fixed route cannot adapt to the diversity of mountainous terrain and the dynamic change of road and bridge structural states. This results in a possible shortage of inspection frequency in some high risk areas, and excessive resources being devoted in some low risk areas, so that the inspection force cannot be distributed reasonably and efficiently.

Disclosure of Invention

In view of the above-mentioned problems, in combination with the first aspect of the present invention, an embodiment of the present invention provides a method for monitoring inspection data applied to early warning of roads and bridges in mountainous areas, the method comprising:

obtaining road and bridge remote sensing monitoring data of a target mountain area road section, wherein the road and bridge remote sensing monitoring data comprise current terrain distribution characteristics, road and bridge structure state parameters and historical disaster records;

based on the current topographic distribution characteristics, calling a topographic characteristic extraction model to carry out area segmentation on the target mountain area road section, and generating a plurality of patrol sub-areas and corresponding topographic characteristic labels;

Generating a dynamic flight path of the unmanned aerial vehicle in the inspection area through a path planning model according to the terrain feature tag and road bridge structure state parameters, wherein the dynamic flight path comprises a plurality of inspection nodes and flight parameters among the inspection nodes;

Controlling an unmanned aerial vehicle to execute a real-time inspection task based on the dynamic flight path, and collecting multi-dimensional monitoring data in the inspection region, wherein the multi-dimensional monitoring data comprises ground surface deformation data, road bridge crack images and slope displacement tracks;

and inputting the multi-dimensional monitoring data and the historical disaster record into a risk early warning model to generate a risk grade signal aiming at a target mountain section and corresponding treatment suggestion information.

In yet another aspect, an embodiment of the present invention further provides a server, including a processor, a machine-readable storage medium, where the machine-readable storage medium is connected to the processor, and the machine-readable storage medium is used to store a program, an instruction, or a code, and the processor is used to execute the program, the instruction, or the code in the machine-readable storage medium, so as to implement the method described above.

Based on the above aspects, the embodiment of the application performs region segmentation and generates the terrain feature label based on the current terrain distribution feature by acquiring the road and bridge remote sensing monitoring data containing the current terrain distribution feature, the road and bridge structure state parameter and the history disaster record, and then generates the unmanned aerial vehicle dynamic flight path by combining the road and bridge structure state parameter, so that the terrain factor, the road and bridge structure and the unmanned aerial vehicle inspection path planning are tightly combined, and the dynamic flight path generated by the method can intelligently adjust the inspection route and the nodes according to the variability of mountain terrain and the difference of road and bridge structure states unlike the traditional fixed path inspection mode. Not only effectively covered all key areas, avoided the blind area of patrolling and examining, can also be to the characteristics and the risk degree in different areas, the rational distribution resource of patrolling and examining improves and patrols and examines efficiency and pertinence. Meanwhile, flight parameters in the dynamic flight path are set, so that the unmanned aerial vehicle can safely, stably and efficiently execute the inspection task in the complex mountain area environment, and the reliability and the practicability of the whole monitoring system are further improved. In a real-time inspection task, multidimensional monitoring data including ground surface deformation data, road and bridge crack images, slope displacement tracks and the like are collected, so that small changes of roads and bridges under the influence of various factors can be accurately captured, and a rich and accurate basis is provided for timely finding potential risks. Finally, the multidimensional monitoring data and the historical disaster record are input into a risk early warning model to generate a risk grade signal and treatment suggestion information, intelligent evaluation and early warning of the risk of a target mountain area road section are realized, the evolution rule and the current actual condition of the mountain area road and bridge disaster are fully considered, the possible disaster risk can be predicted more accurately, and the treatment suggestion with high pertinence and high operability is provided. Compared with the traditional early warning mode, the early warning method has the advantages that the early warning accuracy and timeliness are greatly improved, precious time is striven for timely taking effective precautionary measures, the safe operation of mountain roads and bridges is effectively guaranteed, and disaster loss is reduced.

Drawings

Fig. 1 is a schematic diagram of an execution flow of a method for monitoring patrol data applied to early warning of roads and bridges in a mountain area according to an embodiment of the present invention.

Fig. 2 is a schematic diagram of a server hardware architecture according to an embodiment of the present invention.

Detailed Description

The invention is specifically described below with reference to the accompanying drawings, and fig. 1 is a schematic flow chart of a method for monitoring inspection data applied to mountain road and bridge early warning according to an embodiment of the invention, and the method for monitoring inspection data applied to mountain road and bridge early warning is described in detail below.

Step S110, road bridge remote sensing monitoring data of a target mountain area road section are obtained, wherein the road bridge remote sensing monitoring data comprise current terrain distribution characteristics, road bridge structure state parameters and historical disaster records.

For example, consider a highway segment of a mountain area that is located between mountains, and the surrounding terrain is complex and variable. And acquiring multispectral image data and three-dimensional point cloud data of the target mountain area road section through satellite remote sensing equipment. And obtaining the current topographic distribution characteristic sub-characteristic after the multispectral image data is subjected to the ground object classification treatment. For example, in a certain area, vegetation is dense, vegetation coverage density is high, and other areas are exposed more in rock, and meanwhile, hydrologic distribution characteristics such as rivers can be clearly seen, and the hydrologic distribution characteristics form current topography distribution characteristics. And generating an accurate terrain gradient matrix and an accurate surface relief index for the three-dimensional point cloud data after elevation difference processing. For example, a hillside on one side of a certain section of road has larger gradient and higher surface relief degree.

And extracting the road bridge structure state parameters from the historical maintenance database of the expressway road section. Wherein, there is the slope of certain pier to a certain extent, and its inclination is accurately recorded, and the road surface has the subsidence phenomenon in certain district, and the subsidence rate also is measured and recorded, and slope support structure integrity parameter also can be looked up in the database.

In addition, historical disaster records show that this mountain section has suffered multiple disaster events in the past several years. For example, a landslide event happens at a specific position, so that part of roadbeds are damaged, a bridge crack expansion event is found on a certain bridge, the bridge crack expansion event is repaired but needs to pay close attention, a roadbed collapse event happens on another road section, the disaster events are classified according to types, and the corresponding geographic coordinates and the occurrence time stamp are recorded in detail.

And step S120, calling a topographic feature extraction model to divide the area of the target mountain area road section based on the current topographic distribution feature, and generating a plurality of inspection subareas and corresponding topographic feature labels.

Taking the mountain expressway section mentioned before as an example, the terrain feature extraction model can divide the whole mountain expressway section according to the current terrain distribution features acquired before, such as vegetation coverage density, rock bare area, hydrologic distribution and the like of different areas. For example, a region with dense vegetation, a slow gradient and close to a water source is divided into a patrol sub-region, and the topography of the region is characterized by 'dense vegetation-low gradient-near water'. And for the road sections with exposed rocks and larger gradients, the road sections are divided into another inspection subarea, and the terrain characteristic label is 'rock exposure-high gradient'. Therefore, the whole mountain area road section is divided into a plurality of inspection subareas, each subarea is provided with a unique topographic feature tag, the topographic feature tags accurately reflect the topographic features of the subareas, and an important basis is provided for the subsequent inspection path planning of the unmanned aerial vehicle.

Step S130, generating a dynamic flight path of the unmanned aerial vehicle in the inspection area through a path planning model according to the terrain feature tag and road bridge structure state parameters, wherein the dynamic flight path comprises a plurality of inspection nodes and flight parameters among the inspection nodes.

Taking the mountain expressway section as an example, the path planning model is used for planning the flight path of the unmanned aerial vehicle for different patrol sub-areas divided before and corresponding terrain feature labels and road bridge structure state parameters.

For example, in a dense vegetation-low gradient-near water patrol sub-area, due to the high vegetation coverage density, the terrain gradient matrix shows a low gradient, and according to the related calculation, the minimum safe flight height of the unmanned aerial vehicle in the area is relatively low, and the obstacle avoidance buffer distance is also relatively small. Meanwhile, according to the road bridge structure state parameters, if the inclination angle of a certain bridge pier is smaller and the integrity of the slope support structure is better, when the priority order of the inspection nodes is determined, the priorities of the nodes are relatively lower, and the data acquisition duration of each inspection node is shorter.

For example, for a patrol node of a bridge, the priority is lower due to the better structural state, and the data acquisition time is set to be shorter. Then, the routing nodes are spatially ordered by adopting a path optimization algorithm. Firstly, distributing a corresponding flight height lower limit value and an obstacle avoidance buffer boundary range for each inspection node, wherein the flight height lower limit value is set to be slightly higher than the vegetation height plus a certain safety margin for an inspection node close to the vegetation, and the obstacle avoidance buffer boundary range considers the interference factors of the vegetation. And then sorting the routing inspection nodes according to the priority from high to low, and sorting the routing inspection nodes according to the data acquisition time length from short to long under the condition of the same priority to form a node access sequence with time weight. And then generating a plurality of candidate flight paths between two adjacent routing inspection nodes, for example, from one end routing inspection node to the other end routing inspection node of the bridge, wherein a plurality of different flight paths can be selected, and each candidate flight path can meet the requirements of the obstacle avoidance buffer boundary ranges of the starting point node and the end point node and the flight height.

For each candidate flight path, a composite cost value is calculated. For example, if a candidate flight path is close to an area where landslide events occur, the overlapping proportion of the candidate flight path and the spatial distribution density of disaster points in the historical disaster records is higher, the safety risk cost is higher, and the comprehensive cost value is correspondingly increased. And selecting a path with the lowest comprehensive cost value from the candidate path set as a reference path, and if nodes with the data acquisition time length exceeding a preset threshold value exist in the reference path, optimizing the nodes, such as bridge piers with a certain complex structure, by means of switching node sequences or inserting intermediate transition nodes, so as to generate an optimized target path. And finally, dynamically matching the flight speed of the unmanned aerial vehicle with the trigger frequency of the sensor according to the flight height lower limit value and the obstacle avoidance buffer boundary range of each path section in the target path. For example, in a path segment with a higher flight level, the flight speed may be increased appropriately, while in an area with a smaller range of obstacle avoidance buffer boundaries, the sensor trigger frequency is increased to ensure accuracy of data acquisition.

Meanwhile, the flight parameters are dynamically adjusted based on real-time meteorological data. If the real-time wind speed of the mountain section exceeds a first preset threshold in the inspection process, the unmanned aerial vehicle reduces the flight speed according to a preset deceleration strategy, and increases the obstacle avoidance buffer distance according to a preset obstacle avoidance buffer strategy so as to ensure the flight safety. If the rainfall intensity exceeds a second preset threshold value, an anti-interference mode of the unmanned aerial vehicle is started, the trigger frequency of the sensor is increased according to the first preset compensation parameter, and the accuracy loss can be compensated by increasing the trigger frequency of the sensor because the rainfall possibly affects the data acquisition accuracy. And if the visibility is lower than a third preset threshold, activating an infrared imaging module of the unmanned aerial vehicle to replace a visible light camera, and prolonging the data acquisition time length of the corresponding inspection node according to a second preset compensation parameter so as to acquire more comprehensive and accurate monitoring data. And predicting the flight risk in a future time window according to the change trend of the meteorological data, and if the predicted risk level exceeds a critical value, sending a path re-planning instruction to a navigation control unit of the unmanned aerial vehicle.

In the whole flight path planning, an emergency avoidance node is also inserted into the initial flight path. For example, according to the historical disaster record, in the areas where the landslide events happen for many times, the spatial distribution density is higher, emergency avoidance nodes are configured in a self-adaptive mode in the areas, an adjusted flight path is generated, the adjusted flight path and flight parameters are integrated into a dynamic flight path and are synchronized to a navigation control unit of the unmanned aerial vehicle, a path deviation detection mechanism is preset in the navigation control unit, when the transverse offset of the actual flight path and the dynamic flight path of the unmanned aerial vehicle continuously exceeds the preset proportion of the obstacle avoidance buffer boundary range, local path rescheduling is triggered, and operations such as spatial sequencing are only carried out on the remaining nodes which are not visited, so that the unmanned aerial vehicle can finish the inspection task according to the preset path.

Step S140, controlling the unmanned aerial vehicle to execute a real-time inspection task based on the dynamic flight path, and collecting multi-dimensional monitoring data in the inspection area, wherein the multi-dimensional monitoring data comprises ground surface deformation data, road and bridge crack images and slope displacement tracks.

And on the mountain expressway section, the unmanned aerial vehicle starts to execute the inspection task according to the dynamic flight path. When the unmanned aerial vehicle flies to a certain area, the laser radar sensor scans the surface deformation area. For example, in a road segment where slight landslide signs have occurred, the unmanned aerial vehicle continuously emits laser pulses during the flight, receives reflected signals and calculates pulse round trip time differences, generating an initial point cloud data set. And then denoising the initial point cloud data, removing abnormal points caused by surrounding vegetation shielding or meteorological interference (such as scattering interference caused by foggy days), and generating denoised point cloud data. And then, carrying out elevation correction on the denoised point cloud data based on the terrain gradient matrix acquired before, eliminating the error influence of terrain fluctuation on deformation calculation, and obtaining the current point cloud data. And comparing the current point cloud data with the historical reference data by adopting a time sequence difference algorithm, and calculating the three-dimensional displacement and displacement direction of each monitoring point. If the displacement of a certain area exceeds the early warning threshold, marking the area as a risk area, and distinguishing the risk level of the risk area in the deformation displacement vector diagram by using a color gradient, thereby obtaining the surface deformation data.

Meanwhile, an optical camera is utilized to shoot road and bridge surface images. In the inspection process of a bridge, a surface image of the bridge is shot, the length, width and trend characteristics of a crack corresponding to the image are extracted through an edge detection algorithm, for example, a certain supporting part of the bridge is found to be cracked, the length of the crack is a plurality of centimeters, the width is gradually increased, the trend is along a certain specific direction, and the structural data form a road and bridge crack image.

And calling the multi-band radar of the unmanned aerial vehicle to continuously monitor the side slope. In a certain slope area, the multi-band radar captures the slope displacement track of the rock-soil body in the slope, including a displacement rate curve and a sliding direction index. For example, the displacement rate curve shows that the displacement rate gradually increases over a certain period of time, with the sliding direction being toward the highway bed.

In addition, unmanned aerial vehicle still gather environment humiture, wind speed wind direction and vibration frequency parameter in step. For example, at a certain inspection time, parameters such as the ambient temperature, the humidity, the wind speed and the wind direction, the vibration frequency and the like are accurately collected, and the auxiliary verification information is aligned with the ground deformation data, the road bridge crack image and the slope displacement track according to time sequence and is stored in a distributed database after being bound with corresponding geographic coordinates.

And step S150, inputting the multi-dimensional monitoring data and the historical disaster record into a risk early warning model to generate a risk level signal and corresponding treatment suggestion information aiming at a target mountain section.

And calling a risk early warning model to analyze based on the multi-dimensional monitoring data collected before and the historical disaster record of the mountain section.

For the ground surface deformation data, if the displacement rate of a certain area is continuously increased and the direction is consistent with the historical landslide direction, for example, the ground surface displacement rate under a certain roadbed is continuously increased in a period of time and the displacement direction is the same as the direction of the landslide event, a first-stage early warning signal is triggered.

When the road and bridge crack image is analyzed, if the crack extension direction of a certain key part of the bridge is found to coincide with the main stress direction of the bridge, and the width extension rate exceeds a critical value, for example, the width of the crack is rapidly increased in a short time, so that the safety of the bridge structure is possibly influenced, a secondary early warning signal is triggered.

And calculating a slope stability coefficient according to the sliding direction and the sliding speed of the slope displacement track and combining the real-time rainfall intensity data. For example, the displacement track of a certain slope shows that the sliding direction of the slope is towards the highway, the displacement speed is high, meanwhile, the real-time rainfall intensity is high, and the slope stability coefficient is lower than the safety threshold value by calculation (such as vector decomposition of the sliding direction, comparison with the historical disaster record, searching of the correlation mapping table of the rainfall accumulation and the displacement acceleration interval and the like), so that the three-level early warning signal is triggered.

Differentiated treatment recommendation information is generated based on the pre-alarm signal level. If the primary early warning signal is triggered, measures for emergency closing of road sections can be taken, if the secondary early warning signal is triggered, measures for speed limiting and passing can be taken, and if the secondary early warning signal is triggered, a supporting structure reinforcing scheme can be started.

And finally, pushing the risk level signal and the treatment suggestion information to a road bridge management terminal, and associating corresponding monitoring data traceability links for manual review. For example, the road-bridge management department receives a risk level signal of a certain road section on a terminal as a secondary, the treatment suggestion is limited to pass, and meanwhile detailed monitoring data such as bridge crack images, ground deformation data and the like can be checked through a traceable link so as to perform manual rechecking, and the accuracy of treatment decisions is ensured.

Furthermore, a feedback closed loop mechanism of multi-dimensional monitoring data and treatment recommendation information can be established throughout the process. After the treatment advice information is adopted and executed by the bridge management terminal, the structural response data of the corresponding area is continuously monitored. For example, after a supporting structure reinforcement scheme is implemented on a certain slope, the unmanned aerial vehicle continues to monitor the reinforcement material deformation data and the supporting structure stress distribution data of the area. By comparing the slope displacement track differences before and after treatment, the effective action radius and attenuation coefficient of the treatment measures corresponding to the treatment proposal information are calculated. If the standard deviation of the attenuation coefficient exceeds the preset multiple of the historical average value, determining that the current treatment scheme has design defects, for example, finding that the reinforced slope displacement is still larger and exceeds the expected value, generating an optimization suggestion of material replacement or structural redesign. And inputting the optimization suggestions into a path planning model, increasing the monitoring frequency of the defect area and the accuracy level of the sensor, and restarting a feedback closed-loop mechanism until the actual improvement rate reaches the standard after the new treatment scheme is implemented, so that the monitoring and treatment strategies are continuously optimized, and the safe operation of the road and bridge of the mountain road section is ensured.

Based on the steps, the road and bridge remote sensing monitoring data comprising the current terrain distribution characteristics, road and bridge structure state parameters and the historical disaster records are obtained, the region segmentation is carried out based on the current terrain distribution characteristics, the terrain characteristic labels are generated, then the road and bridge structure state parameters are combined to generate the unmanned aerial vehicle dynamic flight path, the terrain factors, the road and bridge structure and the unmanned aerial vehicle routing inspection path planning are closely combined, and the dynamic flight path generated by the method is different from the traditional fixed path routing inspection mode, so that the routing inspection path and the nodes can be intelligently adjusted according to the variability of mountain terrain and the difference of road and bridge structure states. Not only effectively covered all key areas, avoided the blind area of patrolling and examining, can also be to the characteristics and the risk degree in different areas, the rational distribution resource of patrolling and examining improves and patrols and examines efficiency and pertinence. Meanwhile, flight parameters in the dynamic flight path are set, so that the unmanned aerial vehicle can safely, stably and efficiently execute the inspection task in the complex mountain area environment, and the reliability and the practicability of the whole monitoring system are further improved. In a real-time inspection task, multidimensional monitoring data including ground surface deformation data, road and bridge crack images, slope displacement tracks and the like are collected, so that small changes of roads and bridges under the influence of various factors can be accurately captured, and a rich and accurate basis is provided for timely finding potential risks. Finally, the multidimensional monitoring data and the historical disaster record are input into a risk early warning model to generate a risk grade signal and treatment suggestion information, intelligent evaluation and early warning of the risk of a target mountain area road section are realized, the evolution rule and the current actual condition of the mountain area road and bridge disaster are fully considered, the possible disaster risk can be predicted more accurately, and the treatment suggestion with high pertinence and high operability is provided. Compared with the traditional early warning mode, the early warning method has the advantages that the early warning accuracy and timeliness are greatly improved, precious time is striven for timely taking effective precautionary measures, the safe operation of mountain roads and bridges is effectively guaranteed, and disaster loss is reduced.

In one possible implementation, step S110 includes:

step S111, multispectral image data and three-dimensional point cloud data of a target mountain area road section are obtained through satellite remote sensing equipment.

And step S112, performing ground object classification processing on the multispectral image data, and extracting vegetation coverage density, rock bare area and hydrologic distribution characteristics as sub-characteristics of the current topographic distribution characteristics.

For example, satellites perform omnidirectional scanning imaging of the above-mentioned mountain highway segments from space, and multispectral image data contains rich information. Then, performing ground object classification processing on the multispectral image data to extract sub-features such as vegetation coverage density, rock bare area, hydrologic distribution features and the like. For example, in some sections of a mountain section, vegetation flourishes, and the vegetation coverage density values of the area are calculated through accurate analysis, and the vegetation may have potential influence on the road and bridge structure, for example, the root systems of the vegetation may damage the roadbed. And in other parts, rock bare areas exist, which may suggest that the geological structure stability of the areas is poor, and disasters such as landslide and the like are easy to occur. Meanwhile, the hydrologic distribution characteristics such as the trend of the river and the relative position relation between the river and the road bridge can be clarified through analysis, and if the river is too close to the road bridge, damages such as scouring and the like can be caused to the road bridge foundation.

And step S113, carrying out elevation difference processing on the three-dimensional point cloud data to generate a terrain gradient matrix and an earth surface fluctuation index, and associating the terrain gradient matrix and the earth surface fluctuation index to a corresponding patrol sub-area.

And carrying out elevation difference processing on the acquired three-dimensional point cloud data. In the process, a terrain gradient matrix and an earth surface relief index are generated. For example, in a certain section of a mountain highway, the terrain gradient matrix shows a larger gradient, which means that the section is more challenging to drive or to construct and maintain. The surface relief index also reflects the terrain complexity of the area, such as the greater relief of some hilly areas. And then, the terrain gradient matrix and the surface relief index are related to the corresponding inspection subareas, so that a targeted inspection scheme can be formulated according to different terrain characteristics.

Step S114, a historical maintenance database of the target mountain road section is called, and pier inclination angle, pavement sedimentation rate and slope support structural integrity parameters in the road bridge structural state parameters are extracted.

In the history maintenance database of the target mountain section, a plurality of parameters related to the road bridge structure state are recorded in detail. For example, the pier under a certain bridge of the expressway in the mountain area can obtain the data of the inclination angle of the pier through the periodic measurement of the professional measuring instrument, the numerical value of the inclination angle of the pier reflects the structural stability of the pier, and if the inclination angle is overlarge, the safety of the whole bridge can be affected. The road surface subsidence rate is also an important parameter, and in some road sections, due to the long-term influence of factors such as vehicle load, geological change and the like, the road surface can be subsided, and the accurate road surface subsidence rate can be obtained through analysis of historical data, so that the road surface condition is estimated. In addition, the integrity parameters of the slope support structure are also included, in the slope of the mountain road section, in order to prevent the influence of disasters such as landslide on roads and bridges, the slope support structure is arranged, the detection data of the integrity of the slope support structure are recorded in a database, and the poor integrity means that the protection capability of the slope is possibly reduced, and the risk of landslide exists.

Step S115, classifying the historical disaster records according to disaster types, wherein the disaster types comprise landslide events, bridge crack extension events and roadbed collapse events, and associating geographic coordinates and occurrence time stamps corresponding to the events.

For example, the mountain highway segment has suffered from various disaster types such as landslide event, bridge crack propagation event, and subgrade collapse event. For landslide events, the geographic coordinates of the occurrence of the landslide event are defined, such as on a mountain slope near a certain curve, the longitude and latitude values are accurate, and the occurrence time stamp is recorded, so that the reason of the landslide of the area under specific time and geological conditions is analyzed. And for the bridge crack extension event, determining the specific position and the corresponding geographic coordinate of the bridge crack extension event, and analyzing the crack extension rate and trend through the time stamp. The exact location and time of the subgrade collapse event occurred is also recorded.

In one possible implementation, step S130 includes:

Step S131, calculating the minimum safe flight height and the obstacle avoidance buffer distance of the unmanned aerial vehicle in the corresponding patrol sub-area according to the vegetation coverage density and the terrain gradient matrix in the terrain feature label.

And step S132, determining the priority order of the inspection nodes and the data acquisition time length of each inspection node based on the pier inclination angle and the slope support structure integrity parameter in the road and bridge structure state parameters.

Wherein, the minimum safe flying height H/u min=h_base+α XV+beta.XS ̄

The obstacle avoidance buffer distance d_buffer=γ× (V/v_max) +δ× (Δs/s_ref)

Priority p=θ×w1+ (1-I) ×w2 of the patrol nodes

The data acquisition duration t=t_base× (1+P/p_max)

H_base is the reference safety height, alpha is the weight coefficient of vegetation coverage density V, beta is the correction coefficient of the average slope S ̄ of the terrain slope matrix, gamma is the vegetation density influence factor, V_max is the regional maximum vegetation coverage density, delta is the adjustment coefficient of the slope change rate delta S, S_ref is the slope reference threshold value, theta is the pier inclination angle, I is the integral parameter normalization value of the slope support structure, w1 and w2 are the weight coefficients of the inclination angle and the integral, T_base is the basic acquisition duration, and P_max is the priority maximum.

In this embodiment, taking a certain patrol sub-area in the expressway section of the mountain area as an example, the vegetation coverage density of the area is higher through the previous analysis, and the average gradient of the terrain gradient matrix display is also larger. For the calculation of the minimum safe fly height, the reference safe height h_base is a base set point, here assumed to be a fixed value, for example 100 meters. The vegetation coverage density V is higher, the weight coefficient α is set according to experience and related technical standards, the assumption is 0.5, the average slope S ̄ of the terrain slope matrix is larger, and the correction coefficient β is 0.3. According to the formula of the minimum safe flight height H_min=H_base+alpha×V+beta×S ̄, the minimum safe flight height of the area is calculated to be 120 meters due to the influence of vegetation coverage density and average gradient, so that the unmanned aerial vehicle is ensured not to collide with vegetation or be dangerous due to topography fluctuation in the flight process.

For the obstacle avoidance buffer distance, the maximum vegetation coverage density v_max of the area is assumed to be a specific value, for example, 80%, the vegetation density influence factor gamma is assumed to be 0.2, the gradient change rate deltas is obtained according to the terrain gradient matrix analysis, the gradient reference threshold s_ref is a fixed reference standard value, the gradient reference threshold s_ref is assumed to be 15 degrees, and the adjustment coefficient delta of the gradient change rate is assumed to be 0.1. According to the formula calculation of the obstacle avoidance buffer distance D_buffer=gamma× (V/V_max) +delta× (delta S/S_ref), the obstacle avoidance buffer distance of the area is 5 meters due to the influence of vegetation coverage density and gradient change rate, and the obstacle avoidance buffer distance can provide enough safe buffer space for the unmanned aerial vehicle to avoid obstacles.

Then, at a bridge of the expressway in the mountain area, the inclination angle θ of the bridge pier is measured to be 2 degrees, the normalized value I is 0.8 after the integrity parameter of the slope support structure is evaluated, the weight coefficient w1 of the inclination angle is assumed to be 0.6, the weight coefficient w2 of the integrity is assumed to be 0.4, and the priority of the inspection node is calculated according to the formula of the priority p=θ×w1+ (1-I) ×w2 of the inspection node, so that the priority of the inspection node is 0.52. Assuming that the basic acquisition time length T_base is 10 seconds, the priority maximum value P_max is 1, and calculating according to a formula of the data acquisition time length T=T_base× (1+P/P_max), so that the data acquisition time length of the routing inspection node is 15.2 seconds. In the whole inspection area, different inspection nodes determine the priority sequence and the data acquisition time length according to the corresponding pier inclination angle and the corresponding slope support structure integrity parameter in the calculation mode, and the inspection nodes with larger pier inclination angle or poorer slope support structure integrity are endowed with higher priority and longer data acquisition time length so as to ensure more detailed monitoring on key parts.

Step S133, based on the minimum safe flight height and obstacle avoidance buffer distance of the unmanned aerial vehicle in the corresponding patrol sub-area, the priority sequence of the patrol nodes and the data acquisition time length of each patrol node, adopting a path optimization algorithm to spatially sort the patrol nodes, generating an initial flight path, and dynamically adjusting the flight parameters based on real-time meteorological data, wherein the flight parameters comprise the flight speed, the hovering time and the sensor trigger frequency.

Step S134, inserting an emergency avoidance node into the initial flight path to generate an adjusted flight path, wherein the emergency avoidance node carries out self-adaptive configuration according to the spatial distribution density of disaster types in the historical disaster record.

For example, in mountain highway sections, the spatial distribution density is high in areas where a plurality of landslide events have occurred, and emergency avoidance nodes need to be adaptively configured in these areas. For example, an emergency avoidance node is inserted into a flight path near a road section where landslide often occurs, so that the unmanned aerial vehicle can avoid the dangerous areas in the process of inspection, and an adjusted flight path is generated.

And S135, integrating the adjusted flight path and the flight parameters into the dynamic flight path, synchronizing the dynamic flight path to a navigation control unit of the unmanned aerial vehicle, presetting a path deviation detection mechanism in the navigation control unit, triggering local path rescheduling when the transverse offset of the actual flight path of the unmanned aerial vehicle and the dynamic flight path continuously exceeds a preset proportion of an obstacle avoidance buffer boundary range, and only returning to execute the steps of carrying out spatial sequencing on the routing inspection nodes by adopting a path optimization algorithm for the remaining nodes which are not accessed to generate an initial flight path.

For example, when the transverse offset of the actual flight trajectory and the dynamic flight path of the unmanned aerial vehicle continuously exceeds the preset proportion of the obstacle avoidance buffer boundary range, assuming that the preset proportion is 20%, if the transverse offset of the unmanned aerial vehicle continuously exceeds the proportion, local path rescheduling is triggered, and the steps of spatially sequencing the inspection nodes by adopting a path optimization algorithm only for the remaining nodes which are not accessed are carried out back to generate an initial flight path are carried out, so that the unmanned aerial vehicle can finish the inspection task according to the preset path, and the effective monitoring of the mountain expressway road and bridge structure is ensured.

In one possible implementation, step S133 includes:

step S1331, distributing a corresponding flight height lower limit value and an obstacle avoidance buffer boundary range for each inspection node based on the minimum safe flight height and the obstacle avoidance buffer distance of each inspection sub-region, and generating an inspection node safety parameter set containing a height constraint and a space obstacle avoidance range.

In this embodiment, taking a specific patrol sub-area of a highway in a mountain area as an example, a plurality of patrol nodes exist in the area. One of them is close to hillside and the more node of patrolling and examining of vegetation, and the minimum safe flight height of calculating according to before is 120 meters, then for this node of patrolling and examining the assigned flight height lower limit value of node be 120 meters to ensure that unmanned aerial vehicle can not collide mountain or vegetation because of the altitude is too low when flying through this node. Meanwhile, considering that the obstacle avoidance buffer distance of the area is 5 meters, setting the obstacle avoidance buffer boundary range of the inspection node as an area with the node as the center radius of 5 meters according to the obstacle avoidance buffer distance. And operating each inspection node in the inspection region according to the same method, and finally generating an inspection node safety parameter set containing the height constraint and the space obstacle avoidance range.

Step S1332, sorting the routing inspection nodes according to the priority order and the data acquisition time length of the routing inspection nodes from high to low, and sorting the routing inspection nodes from short to long according to the data acquisition time length under the condition that the priorities are the same, so as to form a node access sequence with time weight.

For example, in the inspection node near a bridge of a highway section in a mountain area, the bridge pier inclination angle is large, the integrity of the slope support structure is relatively poor, and the priority of the inspection node is high according to calculation. And the other inspection node far away from the bridge and having better structural condition has relatively lower priority. For the data acquisition time length, the calculated data acquisition time length of the routing inspection node with high priority is 15.2 seconds, and the data acquisition time length of the routing inspection node with low priority is 10 seconds. According to the sorting rule, firstly, all the routing inspection nodes are arranged from high to low according to the priority, if a plurality of routing inspection nodes have the same priority, for example, if the priority of two routing inspection nodes far away from the main structure is the same, but one of the routing inspection nodes has the data acquisition time length of 8 seconds and the other routing inspection node has the data acquisition time length of 10 seconds, the routing inspection nodes corresponding to the data acquisition time length of 8 seconds are arranged in front according to the data acquisition time length from short to long, and thus a node access sequence with time weight is formed.

Step S1333, based on the sequence of the node access sequences and the inspection node safety parameter set, generating a candidate path set formed by a plurality of candidate flight paths between two adjacent inspection nodes, wherein each candidate flight path meets the condition that the obstacle avoidance buffer boundary range of the starting point node is not overlapped with the obstacle avoidance buffer boundary range of the end point node, and the flight height is always not lower than the higher minimum safety flight height in the two nodes.

For example, in the mountain highway section, two adjacent inspection nodes in the node access sequence, one is a bridge support structure inspection node and the other is a road surface inspection node close to the bridge. For both nodes, a plurality of candidate flight paths are generated between them according to their respective obstacle avoidance buffer boundary ranges. One candidate flight path starts from the bridge support structure inspection node and flies towards the road surface inspection node along a certain angle, the path is required to ensure that the obstacle avoidance buffer boundary range of the starting point node (the bridge support structure inspection node) and the obstacle avoidance buffer boundary range of the end point node (the road surface inspection node) are not overlapped, in the flight process, the flight height is always not lower than the higher minimum safe flight height in the two nodes, for example, the minimum safe flight height of the bridge support structure inspection node is 120 meters, and the minimum safe flight height of the road surface inspection node is 100 meters, so that the flight height of the candidate flight path cannot be lower than 120 meters. According to such rules, a plurality of candidate flight paths are generated between all adjacent routing nodes, forming a candidate path set.

Step S1334, calculating a comprehensive cost value of each candidate flight path according to each candidate flight path, wherein the comprehensive cost value is obtained by weighted summation of flight distance cost, data acquisition time cost and safety risk cost, and the safety risk cost is calculated according to the overlapping proportion of the candidate flight path and the spatial distribution density of disaster points in a historical disaster record.

Taking a candidate flight path from a slope inspection node at one end of a highway section in a mountain area to a bridge inspection node at the other end as an example, firstly calculating the flight distance cost, assuming that the length of the path is 500 meters, and according to a set unit distance cost coefficient (for example, the cost per meter is 0.1 yuan), the flight distance cost is 50 yuan. For the data acquisition time cost, since the sum of the data acquisition time length of the routing inspection node through which the path passes is 30 seconds, the data acquisition time cost is 15 yuan according to the time cost coefficient of each second (for example, the cost of each second is 0.5 yuan). Looking at the safety risk cost, the space distribution density of the landslide event in the candidate flight path and the historical disaster record is overlapped to a certain extent, the overlapping proportion is 30% through detailed calculation, and the safety risk cost is 30 yuan according to the calculation mode of the safety risk cost (for example, the coefficient relation proportional to the overlapping proportion is set, the overlapping proportion corresponds to 10 yuan per 10%). Finally, according to a weighted summation mode (assuming that the cost weight of the flight distance is 0.4, the cost weight of the data acquisition time is 0.3 and the cost weight of the safety risk is 0.3), the comprehensive cost value of the candidate flight path is calculated to be 36 yuan. According to the same method, a composite cost value is calculated for each candidate flight path in the set of candidate paths.

And S1335, selecting a path with the lowest comprehensive cost value from the candidate path set as a reference path, rearranging paths among adjacent nodes of nodes with data acquisition time length exceeding a preset threshold value in the reference path, generating an optimized target path by means of switching node sequences or inserting intermediate transition nodes, reevaluating the comprehensive cost value of the target path until convergence conditions are met, and dynamically matching the flying speed of the unmanned aerial vehicle with the triggering frequency of a sensor according to the flying height lower limit value and the obstacle avoidance buffer boundary range of each path section in the target path after the current optimization, wherein the flying speed is positively correlated with the minimum safe flying height of the path section, and the triggering frequency of the sensor is negatively correlated with the area of the obstacle avoidance buffer boundary range.

Assuming that a data acquisition node with a data acquisition duration of 20 seconds exists in a reference path selected from the candidate path set, and the preset threshold value is 15 seconds, path rearrangement between adjacent nodes is required for the node. If the neighboring node is a node with a data acquisition duration of 10 seconds and a lower priority, the integrated cost value of the new path is reevaluated by exchanging the order of the two nodes. In the process, the node sequence is continuously adjusted or intermediate transition nodes are inserted, and the comprehensive cost value is recalculated after each adjustment until the comprehensive cost value is not reduced any more and reaches a convergence condition. In the optimized target path, for a path section with a higher lower limit value of the flying height, the flying speed and the minimum safe flying height are positively correlated, so that the unmanned aerial vehicle can fly at a relatively higher speed in the path section, for example, a path section with the minimum safe flying height of 150 meters, the flying speed is set to be 20 meters/second, and for a region with a smaller area of the obstacle avoidance buffer boundary, the sensor trigger frequency is required to be increased in the region, for example, a region with the radius of 3 meters of the obstacle avoidance buffer boundary, and the sensor trigger frequency is set to be 5 times per second, so as to ensure that data can be accurately acquired.

Step S1336, detecting whether a target area with the flight direction turning angles exceeding the maximum steering capability of the unmanned aerial vehicle of a continuous preset number of routing inspection nodes exists in the target path, if so, inserting an auxiliary correction node into the target area, determining the position of the auxiliary correction node according to the intersection point of the extension tangents of the obstacle avoidance buffer boundary range, and recalculating the path comprehensive cost value after the auxiliary correction node is inserted until the target path is optimized, and generating the initial flight path.

On the target path of a complex road section of the mountain expressway road section, assuming that the preset number is 3 inspection nodes, the flying direction turning angle of the 3 continuous inspection nodes is found to exceed the maximum steering capability of the unmanned aerial vehicle. And at the moment, determining the position of an auxiliary correction node according to the intersection point of the epitaxial tangents of the obstacle avoidance buffer boundary ranges of the 3 inspection nodes, and inserting the auxiliary correction node at the intersection point position. After insertion, the comprehensive cost value of the path containing the auxiliary correction node is recalculated, then whether the similar problem area exists in the target path is continuously detected, if so, the auxiliary correction node is continuously inserted and the comprehensive cost value is recalculated until the optimization of the target path is completed, and an initial flight path is generated.

Step S1337, acquiring real-time wind speed, rainfall intensity and visibility data of the target mountain road section.

For example, in mountain highway segments, weather monitoring devices are provided that are capable of accurately acquiring real-time weather information for the segment. For example, weather monitoring equipment displays a real-time wind speed of 15 meters/second, a rainfall intensity of 50 millimeters/hour, and a visibility of 50 meters.

Step S1338, if the real-time wind speed exceeds a first preset threshold value, the flying speed of the unmanned aerial vehicle is reduced according to a preset deceleration strategy, the obstacle avoidance buffer distance is increased according to a preset obstacle avoidance buffer strategy, if the rainfall intensity exceeds a second preset threshold value, an anti-interference mode of the unmanned aerial vehicle is started, the triggering frequency of a sensor is increased according to a first preset compensation parameter to compensate the loss of data acquisition precision, if the visibility is lower than a third preset threshold value, an infrared imaging module of the unmanned aerial vehicle is activated to replace a visible light camera, and the data acquisition duration of a corresponding inspection node is prolonged according to the second preset compensation parameter.

If the real-time wind speed exceeds a first preset threshold, for example, the first preset threshold is set to 10 m/s, and since the real-time wind speed exceeds the threshold by 15 m/s, the flight speed of the unmanned aerial vehicle is reduced according to a preset deceleration strategy and the obstacle avoidance buffer distance is increased according to a preset obstacle avoidance buffer strategy. Assuming an original unmanned aerial vehicle having a flight speed of 20 m/s, the flight speed is reduced to 15 m/s according to a preset deceleration strategy (e.g., a1 m/s reduction in the flight speed per 1 m/s of wind speed). Meanwhile, the original obstacle avoidance buffer distance is 5 meters, and according to a preset obstacle avoidance buffer strategy (for example, the obstacle avoidance buffer distance is increased by 1 meter every more than 1 meter/second), the obstacle avoidance buffer distance is increased to 10 meters.

If the rainfall intensity exceeds the second preset threshold, assuming that the second preset threshold is 30 mm/h, starting an anti-interference mode of the unmanned aerial vehicle as the rainfall intensity exceeds the threshold by 50 mm/h, and increasing the triggering frequency of the sensor according to the first preset compensation parameter to compensate the loss of data acquisition precision. For example, the first preset compensation parameter is to increase the sensor trigger frequency by 2 times per second every more than 10 mm/hr of rainfall intensity, and the original sensor trigger frequency is 3 times per second, then the sensor trigger frequency is now increased to 7 times per second, so as to ensure that more accurate data can be acquired in the rainfall environment.

If the visibility is lower than a third preset threshold, assuming that the third preset threshold is 100 meters, and because the visibility is 50 meters lower than the threshold, activating an infrared imaging module of the unmanned aerial vehicle to replace a visible light camera, and prolonging the data acquisition time length of a corresponding inspection node according to a second preset compensation parameter. For example, the second preset compensation parameter is a data acquisition time period of 5 seconds extended for each visibility of less than 50 meters, and for a certain inspection node with an original data acquisition time period of 10 seconds, the data acquisition time period is extended to 15 seconds, so that enough monitoring data can be acquired in a low-visibility environment.

Step S1339, predicting the flight risk in a future time window according to the change trend of the meteorological data, and if the risk level is predicted to exceed a critical value, sending a path re-planning instruction to a navigation control unit of the unmanned aerial vehicle.

For example, through the historical change trend of meteorological data and current data, the wind speed is predicted to continuously increase within 1 hour in the future, the rainfall intensity is also enhanced, the risk level is calculated to exceed a critical value according to a risk assessment model, and a path re-planning instruction is sent to a navigation control unit of the unmanned aerial vehicle at the moment so as to ensure that the unmanned aerial vehicle can safely and effectively fly and collect data in a follow-up inspection task.

In one possible implementation, step S140 includes:

and step S141, scanning the ground surface deformation area through a laser radar sensor of the unmanned aerial vehicle, and generating a deformation displacement vector diagram as the ground surface deformation data.

Step S142, an optical camera is utilized to shoot road and bridge surface images, and the length, width and trend characteristics of the cracks corresponding to the road and bridge surface images are extracted through an edge detection algorithm and are used as structural data of the road and bridge crack images.

For example, when an unmanned aerial vehicle flies over a bridge of a mountain highway, an optical camera captures an image of the bridge surface. The bridge surface image shot by the camera contains various information of the bridge surface, including possible defects such as cracks and the like. Then, the photographed image is processed by an edge detection algorithm. The edge detection algorithm can identify the edge of an object in the bridge surface image, and can accurately determine the boundary of a crack of the bridge surface. By analyzing these boundary information, the length of the crack is calculated, for example, the distance from one end of the crack to the other is measured to be several centimeters. Meanwhile, the width of the crack can be determined, and the value of the width of the crack can be obtained by combining known image scale through pixel analysis of the edge of the crack. And determining the trend characteristics of the crack on the actual bridge surface according to the trend of the crack in the image, such as extending along the transverse direction or the longitudinal direction of the bridge. These crack length, width and strike characteristics constitute the structured data of the road bridge crack image.

And S143, invoking a multi-band radar of the unmanned aerial vehicle to continuously monitor the side slope, and capturing a side slope displacement track of a rock-soil body in the side slope, wherein the side slope displacement track comprises a displacement rate curve and a sliding direction index.

In the slope area of the expressway section in the mountain area, the multi-band radar continuously transmits radar waves to the slope and receives reflected waves. The reflected wave will change due to the movement of the rock and soil mass inside the slope. Through analysis of the reflected waves, the multi-band radar can capture displacement information of rock and soil bodies in the slope. With the lapse of time, the displacement information is continuously collected, and a displacement rate curve can be drawn. For example, over a period of time, the rate of displacement of the rock mass at a depth may begin to be lower and then gradually increase, this change being manifested in the displacement rate curve. Meanwhile, according to the information such as the direction change of radar wave reflection, the sliding direction index of the rock-soil body can be determined, for example, whether the rock-soil body slides towards the road direction or slides towards the inside of a hillside, and the displacement rate curve and the sliding direction index form a slope displacement track.

And S144, synchronously acquiring parameters of environmental temperature and humidity, wind speed and wind direction and vibration frequency, and taking the parameters as auxiliary verification information of the multi-dimensional monitoring data.

And S145, aligning the surface deformation data, the road and bridge crack images and the slope displacement track according to time sequences, binding the surface deformation data, the road and bridge crack images and the slope displacement track with corresponding geographic coordinates, and storing the surface deformation data, the road and bridge crack images and the slope displacement track in a distributed database.

In the whole inspection process, equipment such as temperature and humidity sensors, anemometers, vibration sensors and the like carried by the unmanned aerial vehicle work simultaneously. In the inspection area of the mountain expressway, the temperature and humidity sensor can accurately measure the temperature and humidity of the environment, for example, the temperature of the environment is 25 ℃ at a certain moment, and the humidity is 60%. The anemometer can detect the current wind speed and direction, for example, the wind speed is 10 m/s, and the wind direction is southeast. The vibration sensor may detect the vibration frequency of the drone itself or the surrounding environment, assuming that the detected vibration frequency is 100 hertz. The environmental temperature and humidity, wind speed and wind direction and vibration frequency parameters are aligned with the ground surface deformation data, road and bridge crack images and slope displacement tracks according to time sequences. For example, when the surface deformation data at a certain moment is acquired, the environmental parameters at the moment are recorded at the same time, so that all the data are ensured to be corresponding in time. And binding the data with corresponding geographic coordinates, for example, coordinates of a certain section of mountain expressway corresponding to certain surface deformation data, and storing the bound data into a distributed database to facilitate subsequent query, analysis and processing.

In one possible implementation, step S141 includes:

Step 1411, continuously transmitting laser pulses in the flight process of the unmanned aerial vehicle, receiving the reflected signals and calculating pulse round trip time difference to generate an initial point cloud data set.

In a certain inspection area of the mountain highway, there is a surface area where a slight landslide sign has occurred. When the unmanned plane flies above the area, the laser radar continuously emits laser pulses in the flying process, the laser pulses are reflected back after being emitted to the ground surface, and the sensor receives reflected signals and accurately calculates pulse round-trip time difference. For example, after a certain laser pulse is emitted, the laser pulse returns after a period of time, and according to the propagation speed of the laser in the air, the distance information from the monitoring point corresponding to the pulse to the unmanned aerial vehicle can be obtained by calculating the round trip time difference. A large amount of such distance information is collected by multiple laser pulse transmissions and receptions of the entire surface deformation region, thereby generating an initial point cloud data set.

Step S1412, denoising the initial point cloud data, removing abnormal points caused by vegetation shielding or meteorological interference, and generating denoised point cloud data.

In this embodiment, this initial point cloud dataset contains information for a large number of points of the surface area, but there may be some outliers. These outliers may be due to vegetation shadows or weather disturbances. For example, in areas where there are large trees, the laser pulses may scatter or reflect multiple times as they pass through the tree branches and leaves, resulting in inaccurate distance information being calculated, and interference with the laser pulses propagation under meteorological conditions, such as fog or rainfall. Therefore, denoising is needed to be carried out on the initial point cloud data, and abnormal points caused by vegetation shielding or meteorological interference are removed, so that denoised point cloud data are generated.

Step S1413, carrying out elevation correction on the denoised point cloud data based on the terrain gradient matrix to obtain current point cloud data, and eliminating error influence of terrain fluctuation on deformation calculation.

And step 1414, comparing the current point cloud data with the historical reference data by adopting a time sequence difference algorithm, and calculating the three-dimensional displacement and displacement direction of each monitoring point.

And step S1415, marking the area with the displacement exceeding the early warning threshold value as a risk area, and distinguishing the risk level of the risk area in the deformation displacement vector diagram by using a color gradient.

In detail, the history reference data is data previously collected and processed in the same area as a reference for comparison. The three-dimensional displacement can be calculated by comparing the coordinate information of each monitoring point in the current point cloud data and the historical reference data. For example, the coordinates of a certain monitoring point in the current data are different from the coordinates of a certain monitoring point in the historical reference data in the x, y and z directions respectively, and the difference is the three-dimensional displacement of the monitoring point. Meanwhile, the displacement direction can be determined according to the change trend of the coordinates, such as towards a specific geographic direction. If the displacement of a certain area exceeds the early warning threshold value, the area is marked as a risk area, and the risk level of the risk area is distinguished by a color gradient in the deformation displacement vector diagram. For example, a region with a small displacement amount is represented by light blue, a region with a large displacement amount near a dangerous value is represented by yellow, and a region with a large displacement amount exceeding the dangerous value is represented by red, so that a deformation displacement vector diagram is generated as the ground deformation data.

For example, step S1413 includes:

Step S1413-1, acquiring denoised point cloud data, and extracting three-dimensional coordinate information of each monitoring point in the denoised point cloud data, wherein the three-dimensional coordinate information comprises a horizontal coordinate and a vertical elevation value.

Step S1413-2, the terrain gradient matrix is called, the terrain gradient matrix contains the terrain gradient distribution information of the target mountain section, the terrain gradient distribution information is composed of a plurality of gradient units, and each gradient unit corresponds to a geographic area.

And step 1413-3, performing space matching on the horizontal coordinate of each monitoring point in the denoised point cloud data and the gradient unit in the terrain gradient matrix, and determining the gradient unit corresponding to each monitoring point.

And step S1413-4, calculating a terrain fluctuation correction amount of each monitoring point according to the terrain gradient distribution information of the gradient unit, wherein the terrain fluctuation correction amount is used for eliminating the influence of the terrain gradient on the vertical elevation value.

Step S1413-5, applying the topographic relief correction amount to the vertical elevation value of each monitoring point in the denoised point cloud data to generate a corrected vertical elevation value, wherein the corrected vertical elevation value eliminates the error influence of topographic relief on deformation calculation.

And step S1413-6, recombining the corrected vertical elevation value and the horizontal coordinate in the denoised point cloud data to generate corrected current point cloud data.

In this embodiment, although the denoised point cloud data is relatively accurate, due to the existence of the topography relief, an error effect may be generated on the deformation calculation. To eliminate such an influence, elevation correction of the denoised point cloud data is required based on the terrain gradient matrix. Firstly, denoising point cloud data, and extracting three-dimensional coordinate information of each monitoring point, wherein the three-dimensional coordinate information comprises horizontal coordinates and vertical elevation values. The previously acquired terrain gradient matrix is then retrieved, which contains terrain gradient distribution information for the target mountain section, which information is composed of a plurality of gradient units, each gradient unit corresponding to a geographical area. And performing space matching on the horizontal coordinates of each monitoring point in the denoised point cloud data and gradient units in the terrain gradient matrix, and determining the gradient unit corresponding to each monitoring point. For example, if the horizontal coordinate of a monitoring point is within a particular grade cell, then it is determined that the monitoring point corresponds to the grade cell. Then, according to the information of the terrain gradient distribution of the gradient unit, the terrain fluctuation correction amount of each monitoring point is calculated, and the terrain fluctuation correction amount is specially used for eliminating the influence of the terrain gradient on the vertical elevation value. For example, if the slope of a slope unit is large, the vertical elevation value of the corresponding monitoring point may have a large deviation due to the slope of the terrain, and the calculated terrain relief correction amount may be used to correct the deviation. The terrain relief correction amount is applied to the vertical elevation value of each monitoring point in the denoised point cloud data, so that a corrected vertical elevation value is generated, which has eliminated the error effect of the terrain relief on the deformation calculation. And finally, recombining the corrected vertical elevation value with the horizontal coordinates in the denoised point cloud data to generate corrected current point cloud data.

In one possible implementation, step S150 includes:

And step S151, calling the risk early warning model, matching the displacement rate in the surface deformation data with a landslide event in the historical disaster record, and triggering a primary early warning signal if the displacement rate is continuously increased and the direction is consistent with the historical landslide direction.

When analyzing the surface deformation data, the displacement rate is focused on, and the displacement rate is matched with landslide events in the historical disaster records. For example, the surface deformation data of a certain section of expressway in a mountain area shows that the displacement rate of a certain specific area shows a continuously increasing trend. It is known from a historical disaster record query of the area that a landslide event has occurred before, and the displacement direction monitored at this time is consistent with the historical landslide direction. This indicates that this region has extremely high landslide risk, satisfies the condition that triggers the first-level early warning signal. This situation may be due to the fact that the geological structure of the area is fragile, and in addition, the area may be affected by rain wash or groundwater level change in the near term, so that the stability of the earth surface soil body is gradually reduced, and further displacement change characteristics similar to that of a historical landslide appear.

And step S152, analyzing trend characteristics of the road and bridge crack image, and triggering a secondary early warning signal if the crack extension direction coincides with the main stress direction of the bridge and the width extension rate exceeds a critical value.

In the inspection of the mountain expressway bridge, the road and bridge crack image shot by the optical camera is processed to obtain the relevant characteristic information of the crack. If the extending direction of a certain crack is found to coincide with the main stress direction of the bridge, the developing direction of the crack is consistent with the least favorable stress direction of the bridge structure, and the structural safety of the bridge is seriously threatened. Meanwhile, if the crack width expansion rate exceeds a critical value, for example, in the process of continuous inspection for several times, the crack width is found to increase at a faster speed and exceeds a critical value set according to bridge structure design and safety standards, so that the damage of the bridge structure is accelerated, and a secondary early warning signal is triggered at the moment. The development of such cracks may be due to long-term vehicle loading, temperature changes, or material aging, and the coincidence of the cracks with the direction of principal stresses accelerates the failure process of the bridge structure.

And step S153, calculating a slope stability coefficient according to the sliding direction and the sliding speed of the slope displacement track and combining the real-time rainfall intensity data, and triggering a three-level early warning signal if the slope stability coefficient is lower than a safety threshold value.

Step S154, differentiated treatment advice information is generated based on the early warning signal level, the treatment advice information including emergency closed road segments, speed limit traffic or starting a support structure reinforcement scheme.

For example, when the first-level early warning signal is triggered, due to extremely high landslide risk, measures for emergency closing the road section are required to ensure the driving safety of the expressway in the mountain area and the safety of the road facilities. This means that the vehicle is immediately prevented from entering the dangerous area, a roadblock and warning sign are set, and relevant departments are informed to perform emergency treatment, such as geological investigation and reinforcement treatment on landslide areas.

If the secondary early warning signal is triggered, the bridge structure has a large potential safety hazard, but the degree of immediately interrupting traffic is not reached. At this time, the treatment advice information of speed limit traffic is adopted. Speed limit marks are arranged at two ends of the bridge, the speed of a vehicle passing through the bridge is reduced, the impact force of the vehicle load on the bridge is reduced, and professionals are arranged to further inspect and evaluate the bridge in detail, so that a maintenance and reinforcement plan is formulated.

When the three-level early warning signal is triggered, a supporting structure reinforcing scheme is started aiming at the slope stability problem. This may include measures such as reinforcing the retaining wall of the slope, increasing the number and length of anchor rods or anchor cables, protecting the slope surface, etc., to improve the stability of the slope and prevent the slope from sliding.

Step S155, pushing the risk level signal and the treatment suggestion information to a bridge management terminal, and associating corresponding monitoring data traceability links for manual review.

For example, after the road bridge management terminal receives the information that the risk level of a certain road section is first-level and the treatment proposal is an emergency closed road section, detailed monitoring data such as a displacement rate change curve and displacement direction information in surface deformation data, related records of historical landslide events and the like can be checked through the associated monitoring data traceability link. Therefore, bridge management personnel can manually review the early warning information and the treatment advice, ensure the accuracy and rationality of decision making, and adjust or supplement the treatment advice according to actual conditions when necessary.

In one possible implementation, step S153 includes:

Step S1531, vector-decomposing the sliding direction of the slope displacement track to generate time series data of horizontal displacement component and vertical displacement component.

For example, in monitoring a certain slope of a highway in a mountain area, a multi-band radar captures the displacement track of a rock-soil body in the slope, and the displacement change conditions in the horizontal direction and the vertical direction in a period of time are obtained through a vector decomposition technology. The horizontal displacement component reflects the transverse movement trend of the rock-soil body in the slope plane, and the vertical displacement component reflects the sedimentation or rising trend of the rock-soil body.

Step S1532, performing similarity matching on the horizontal displacement component and a landslide event displacement mode of a corresponding geographic coordinate in the historical disaster record, and screening out a reference event set with displacement trend similarity exceeding a matching threshold.

In the history disaster records of mountain expressways, a detailed landslide event record is provided for each side slope area, including a displacement mode of a rock-soil body when landslide occurs. And screening out a reference event set with the similarity of the displacement trend exceeding a matching threshold value by comparing the horizontal displacement component of the current side slope with the displacement mode in the history record. This matching threshold is determined based on a large amount of historical data and engineering experience to ensure that the reference events that are screened have a high reference value.

Step S1533, extracting rainfall intensity time sequence data before each landslide event in the reference event set, and establishing an association mapping table of rainfall accumulation and displacement acceleration interval.

The rainfall intensity time sequence data records the change condition of rainfall intensity in a period of time before landslide event. Through analysis of the data, a correlation mapping table of rainfall accumulation and displacement acceleration interval is established. For example, in a certain reference landslide event, it is found that when the rainfall accumulation reaches a certain value, the displacement rate of the side slope rock-soil body starts to be significantly accelerated, and enters a displacement acceleration section. By analyzing a plurality of reference events, the relation between different rainfall accumulation amounts and displacement acceleration intervals can be established.

Step S1534, calculating the current rainfall accumulation according to the real-time rainfall intensity data, and searching a critical rainfall threshold overlapping with the current displacement acceleration interval in the association mapping table.

In the process of inspection, real-time rainfall intensity data are acquired through meteorological monitoring equipment arranged on a highway section in a mountain area, and accumulation calculation is carried out according to a certain time interval, so that the current rainfall accumulation is obtained. And then searching a critical rainfall threshold value overlapped with the current displacement acceleration interval in the association mapping table. Assuming that the displacement of the current slope is in a certain acceleration interval, searching the association mapping table to obtain a corresponding critical rainfall threshold value of 100 mm.

Step S1535, generating a dynamic correction factor of the slope stability coefficient based on the ratio of the change rate of the vertical displacement component to the critical rainfall threshold.

For example, if the rate of change of the vertical displacement component is 5 mm/hr and the critical rainfall threshold is 100 mm, the dynamic correction factor is 5/100=0.05. The dynamic correction factor reflects the change degree of the vertical displacement of the slope relative to the critical rainfall condition under the current rainfall condition.

And step S1536, carrying out weighted fusion on the dynamic correction factors and the normalized values of the integrity parameters of the side slope support structure, and outputting the comprehensive side slope stability coefficients.

Assuming that the normalized value of the structural integrity parameter of the side slope support is 0.8 after evaluation, carrying out weighted fusion calculation according to a set weighting coefficient (for example, the dynamic correction factor weight is 0.3 and the structural integrity parameter weight of the side slope support is 0.7), so as to obtain the comprehensive side slope stability coefficient of 0.8x0.7+0.05x0.3=0.575. And if the comprehensive slope stability coefficient is lower than the safety threshold value, triggering a three-level early warning signal. This shows that under the current slope state, due to the influence of factors such as rainfall, rock-soil mass displacement and the like, the stability of the slope is in a dangerous range, and corresponding measures need to be taken.

In one possible embodiment, the method further comprises:

step S210, establishing a feedback closed-loop mechanism of the multi-dimensional monitoring data and treatment recommendation information.

Step S220, after the treatment proposal information is adopted and executed by the bridge management terminal, continuously monitoring the structural response data of the corresponding area, comparing the structural response data with the expected treatment effect, and triggering a treatment scheme optimization instruction if the comparison analysis result represents that the actual improvement rate is lower than the set proportion of the expected value.

For example, after a treatment proposal of a supporting structure reinforcing scheme is adopted for a certain slope, reinforcing material deformation data and supporting structure stress distribution data of a treatment area are synchronously collected when the unmanned aerial vehicle executes a patrol task. In the reinforced slope area, deformation data of the reinforcing material, such as the elongation of the anchor rod or the inclination change of the retaining wall, is acquired through a sensor arranged in the reinforcing material. Meanwhile, stress distribution data of the supporting structure are obtained through stress sensors arranged in the supporting structure, and the size and distribution condition of internal stress of the supporting structure are known.

These structural response data are then analyzed in comparison to expected treatment effects. The expected treatment effect is preset according to the treatment scheme, for example, the displacement rate of the slope is expected to be reduced to be within a certain safety range within a certain period of time, the deformation of the reinforcing material is expected to be kept within a design allowable range, and the stress distribution of the supporting structure is expected to conform to the structural mechanics principle and the like. If the comparison analysis results indicate that the actual improvement rate is lower than the set proportion of the expected value, for example, the expected slope displacement rate is reduced by 80%, but is reduced by only 50% in practice, a treatment protocol optimization instruction is triggered.

Step S230, parameter weights of the risk early warning model are adjusted based on the treatment scheme optimization instruction, and after the emergency avoidance node configuration in the path planning model is updated, optimized model parameters and configuration information are synchronized to all edge computing units of the in-service unmanned aerial vehicle, so that online iteration upgrading of the monitoring strategy is realized.

For example, if the current treatment scheme is found to have poor control effect on the slope displacement, the weight of the relation between the slope displacement rate and the rainfall intensity in the risk early-warning model may be unreasonably set. And adjusting the weight of related parameters in the risk early warning model, such as increasing the weight of the influence of rainfall intensity on the slope stability. Meanwhile, due to the fact that unstable states of the side slopes possibly influence the routing inspection path planning of the unmanned aerial vehicle, emergency avoidance node configuration in the path planning model is updated, for example, more emergency avoidance nodes are added near the side slopes or positions of existing avoidance nodes are adjusted. And then synchronizing the optimized model parameters and configuration information to all edge computing units of the in-service unmanned aerial vehicle, so that the unmanned aerial vehicle can carry out more effective inspection according to a new monitoring strategy in a subsequent inspection task.

For example, the implementation of the feedback closed loop mechanism includes:

Step S211, synchronously collecting reinforcing material deformation data and support structure stress distribution data of a treatment area when the unmanned aerial vehicle executes a patrol task.

Step S212, by comparing the slope displacement trajectory differences before and after treatment, the effective radius of action and attenuation coefficient of the treatment measure corresponding to the treatment advice information are calculated.

In step S213, if the attenuation coefficient exceeds the standard deviation of the preset multiple of the historical average, it is determined that the current treatment scheme has a design defect, and an optimization suggestion of material replacement or structural redesign is generated.

Step S214, inputting the optimization suggestion into the path planning model, and increasing the monitoring frequency of the defect area and the sensor precision level.

And step S215, when a new treatment scheme is implemented, restarting the feedback closed-loop mechanism until the actual improvement rate reaches the standard.

For example, slope displacement trajectory data is acquired before and after reinforcement treatment of the slope. By comparison, the slope displacement is well controlled within a certain range from the reinforcing structure, and the range is the effective radius of action of the treatment measures. The control effect of the treatment measures on the slope displacement may gradually decrease over time, and the attenuation coefficient is calculated by analyzing the change trend of the displacement trajectory data. If the implemented attenuation coefficient exceeds the standard deviation of the preset multiple of the historical average value, for example, the historical average attenuation coefficient is 0.1, the standard deviation is 0.02, the currently calculated attenuation coefficient is 0.15, and the standard deviation exceeds 1.5 times, the current treatment scheme is judged to have design defects, and optimization suggestions of material replacement or structural redesign are generated.

And inputting the optimization suggestion into a path planning model, and increasing the monitoring frequency of the defect area and the accuracy level of the sensor. For example, if it is determined to be a problem with the reinforcement material, optimization advice for material replacement is entered into the path planning model such that the drone increases the frequency of monitoring the reinforcement material for the area during subsequent inspection, such as from original once per day to three times per day. Meanwhile, the precision level of the sensor is improved, so that deformation data of the reinforcing material and stress distribution data of the supporting structure can be acquired more accurately. After the new treatment scheme is implemented, the feedback closed-loop mechanism is restarted until the actual improvement rate reaches the standard, so that the slope of the expressway section in the mountain area is ensured to be in a safe and stable state, and the safe operation of the whole road is ensured.

Fig. 2 shows a schematic diagram of exemplary hardware and software components of a server 100 that may implement the concepts of the present application provided by some embodiments of the present application. For example, processor 120 may be used on server 100 and to perform the functions of the present application.

The server 100 may be a general-purpose server or a special-purpose server, and both servers may be used to implement the inspection data monitoring method of the present application applied to mountain road and bridge early warning. Although only one server is shown, the functionality described herein may be implemented in a distributed fashion across multiple similar platforms for convenience to balance processing loads.

For example, the server 100 may include a network port 110 connected to a network, one or more processors 120 for executing program instructions, a communication bus 130, and various forms of storage media 140, such as magnetic disk, ROM, or RAM, or any combination thereof. By way of example, server 100 may also include program instructions stored in ROM, RAM, or other types of non-transitory storage media, or any combination thereof. The method of the present application may be implemented in accordance with these program instructions. The server 100 also includes an Input/Output (I/O) interface 150 between a computer and other Input/Output devices.

For ease of illustration, only one processor is depicted in server 100. It should be noted, however, that the server 100 of the present application may also include a plurality of processors, and thus the steps performed by one processor described in the present application may also be performed jointly by a plurality of processors or separately. For example, if the processor of the server 100 performs the steps a and B, it should be understood that the steps a and B may be performed by two different processors together or performed separately in one processor. For example, the first processor performs step a, the second processor performs step B, or the first processor and the second processor together perform steps a and B.

In addition, the embodiment of the invention also provides a readable storage medium, wherein computer executable instructions are preset in the readable storage medium, and when a processor executes the computer executable instructions, the inspection data monitoring method applied to mountain road and bridge early warning is realized.

It should be noted that in order to simplify the presentation of the disclosure and thereby aid in understanding one or more embodiments of the invention, various features are sometimes grouped together in a single embodiment, figure, or description thereof.

Claims (10)

1. A patrol data monitoring method for early warning of roads and bridges in mountainous areas, characterized in that the method comprises: Acquire remote sensing monitoring data of roads and bridges in target mountainous road sections, wherein the remote sensing monitoring data of roads and bridges includes current terrain distribution characteristics, road and bridge structure status parameters, and historical disaster records; Based on the current terrain distribution characteristics, calling the terrain feature extraction model to perform regional segmentation on the target mountainous road section, and generating multiple inspection sub-areas and corresponding terrain feature labels; According to the terrain feature labels and the road and bridge structure state parameters, a dynamic flight path of the UAV in the inspection sub-area is generated through a path planning model, wherein the dynamic flight path includes a plurality of inspection nodes and flight parameters between the inspection nodes; Based on the dynamic flight path, the UAV is controlled to perform real-time inspection tasks, and multi-dimensional monitoring data in the inspection sub-area is collected, wherein the multi-dimensional monitoring data includes surface deformation data, road and bridge crack images, and slope displacement trajectories; The multi-dimensional monitoring data and the historical disaster records are input into a risk warning model to generate a risk level signal and corresponding handling suggestion information for the target mountainous road section. 2. The inspection data monitoring method for early warning of mountainous roads and bridges according to claim 1 is characterized in that the acquisition of remote sensing monitoring data of roads and bridges in target mountainous road sections comprises: Acquire multispectral image data and three-dimensional point cloud data of the target mountainous road section through satellite remote sensing equipment; Performing ground object classification processing on the multispectral image data, extracting vegetation coverage density, rock exposed area and hydrological distribution characteristics as sub-features of the current terrain distribution characteristics; Performing elevation difference processing on the three-dimensional point cloud data to generate a terrain slope matrix and a surface undulation index, and associating the terrain slope matrix and the surface undulation index with corresponding inspection sub-areas; Retrieving the historical maintenance database of the target mountainous road section, extracting the pier inclination angle, road surface settlement rate and slope support structure integrity parameters from the road and bridge structure status parameters; The historical disaster records are classified according to disaster types, including landslide events, bridge crack expansion events and roadbed collapse events, and the geographical coordinates and occurrence timestamps corresponding to each event are associated. 3. The inspection data monitoring method for early warning of road and bridge in mountainous areas according to claim 1 is characterized in that the dynamic flight path of the UAV in the inspection sub-area is generated by a path planning model according to the terrain feature labels and the road and bridge structure state parameters, comprising: According to the vegetation coverage density and terrain slope matrix in the terrain feature label, the minimum safe flight altitude and obstacle avoidance buffer distance of the UAV in the corresponding inspection sub-area are calculated; Based on the pier inclination angle and slope support structure integrity parameters in the road and bridge structure status parameters, determine the priority order of the inspection nodes and the data collection time of each inspection node; Based on the minimum safe flight altitude and obstacle avoidance buffer distance of the drone in the corresponding inspection sub-area, as well as the priority order of the inspection nodes and the data collection time of each inspection node, a path optimization algorithm is used to spatially sort the inspection nodes, generate an initial flight path, and dynamically adjust the flight parameters based on real-time meteorological data, including flight speed, hovering time and sensor trigger frequency; Inserting emergency avoidance nodes into the initial flight path to generate an adjusted flight path, wherein the emergency avoidance nodes are adaptively configured according to the spatial distribution density of disaster types in historical disaster records; Integrate the adjusted flight path and the flight parameters into the dynamic flight path, and synchronize the dynamic flight path to the navigation control unit of the UAV, preset a path deviation detection mechanism in the navigation control unit, and trigger local path replanning when the lateral offset between the actual flight trajectory of the UAV and the dynamic flight path continues to exceed a preset ratio of the obstacle avoidance buffer boundary range, and return to execute the step of spatially sorting the inspection nodes using a path optimization algorithm to generate an initial flight path only for the remaining nodes that have not been visited; Wherein, the minimum safe flight altitude H_min=H_base+α×V+β×S̄; The obstacle avoidance buffer distance D_buffer=γ×(V/V_max)+δ×(ΔS/S_ref); The priority of the inspection node P=θ×w ₁ +(1-I)×w _2; The data collection duration T=T_base×(1+P/P_max); H_base is the benchmark safety height, α is the weight coefficient of the vegetation coverage density V, β is the correction coefficient of the average slope S̄ of the terrain slope matrix, γ is the influencing factor of vegetation density, V_max is the maximum vegetation coverage density in the region, δ is the adjustment coefficient of the slope change rate ΔS, S_ref is the slope reference threshold, θ is the inclination angle of the pier, I is the normalized value of the integrity parameter of the slope support structure, _w1 and _w2 are the weight coefficients of the inclination angle and integrity respectively, T_base is the basic collection time, and P_max is the maximum priority value. 4. The inspection data monitoring method for mountainous road and bridge early warning according to claim 3 is characterized in that the step of spatially sorting the inspection nodes using a path optimization algorithm based on the minimum safe flight altitude and obstacle avoidance buffer distance of the drone in the corresponding inspection sub-area, the priority order of the inspection nodes and the data collection time of each inspection node to generate an initial flight path comprises: Based on the minimum safe flight altitude and obstacle avoidance buffer distance of each inspection sub-area, a corresponding flight altitude lower limit and obstacle avoidance buffer boundary range are assigned to each inspection node, and a set of inspection node safety parameters including altitude constraints and spatial obstacle avoidance ranges is generated; According to the priority order of the inspection nodes and the data collection time, the inspection nodes are sorted from high to low according to the priority, and when the priorities are the same, they are sorted from short to long according to the data collection time to form a node access sequence with time weight; Based on the order of the node access sequence and the inspection node safety parameter set, a candidate path set consisting of multiple candidate flight paths is generated between two adjacent inspection nodes, and each candidate flight path satisfies that the obstacle avoidance buffer boundary range of the starting node and the obstacle avoidance buffer boundary range of the end node do not overlap, and the flight altitude is always not lower than the higher minimum safe flight altitude of the two nodes; For each candidate flight path, a comprehensive cost value of the candidate flight path is calculated, wherein the comprehensive cost value is obtained by weighted summation of flight distance cost, data collection time cost and safety risk cost, wherein the safety risk cost is calculated according to the overlap ratio between the candidate flight path and the spatial distribution density of disaster points in historical disaster records; The path with the lowest comprehensive cost value is selected from the candidate path set as the reference path, and the paths between adjacent nodes are rearranged for nodes whose data collection time exceeds a preset threshold in the reference path, and an optimized target path is generated by exchanging the node order or inserting an intermediate transition node, and the comprehensive cost value of the target path is re-evaluated until the convergence condition is reached. In the currently optimized target path, the flight speed and sensor trigger frequency of the UAV are dynamically matched according to the flight altitude lower limit value and obstacle avoidance buffer boundary range of each path segment in the target path, wherein the flight speed is positively correlated with the minimum safe flight altitude of the path segment, and the sensor trigger frequency is negatively correlated with the area of the obstacle avoidance buffer boundary range; Detect whether there is a target area in the target path where the flight direction turning angles of a preset number of consecutive inspection nodes exceed the maximum steering capability of the drone. If so, insert an auxiliary correction node in the target area, the position of the auxiliary correction node is determined according to the intersection of the extended tangent of the obstacle avoidance buffer boundary range, and recalculate the comprehensive cost value of the path after inserting the auxiliary correction node until the target path optimization is completed to generate the initial flight path; The dynamically adjusting the flight parameters based on real-time meteorological data includes: Obtain real-time wind speed, rainfall intensity and visibility data for target mountainous road sections; If the real-time wind speed exceeds the first preset threshold, the flight speed of the UAV is reduced according to the preset deceleration strategy and the obstacle avoidance buffer distance is increased according to the preset obstacle avoidance buffer strategy; If the rainfall intensity exceeds the second preset threshold, the anti-interference mode of the drone is enabled, and the sensor trigger frequency is increased according to the first preset compensation parameter to compensate for the loss of data collection accuracy; If the visibility is lower than the third preset threshold, the infrared imaging module of the drone is activated to replace the visible light camera, and the data collection time of the corresponding inspection node is extended according to the second preset compensation parameter; The flight risk in the future time window is predicted based on the changing trend of meteorological data. If the predicted risk level exceeds the critical value, a path replanning instruction is sent to the navigation control unit of the drone. 5. The inspection data monitoring method for mountain road and bridge early warning according to claim 2 is characterized in that the control of the drone based on the dynamic flight path to perform real-time inspection tasks and collect multi-dimensional monitoring data in the inspection sub-area includes: Scanning the surface deformation area by using a laser radar sensor of the drone to generate a deformation displacement vector map as the surface deformation data; Using an optical camera to capture a road bridge surface image, and using an edge detection algorithm to extract crack length, width, and direction features corresponding to the road bridge surface image as structured data of the road bridge crack image; The multi-band radar of the UAV is used to continuously monitor the slope and capture the slope displacement trajectory of the rock and soil inside the slope, which includes the displacement rate curve and the sliding direction index; Synchronously collect environmental temperature and humidity, wind speed and direction, and vibration frequency parameters as auxiliary verification information for the multi-dimensional monitoring data; The surface deformation data, road and bridge crack images and slope displacement trajectories are aligned in time series, bound with corresponding geographic coordinates and stored in a distributed database. 6. The inspection data monitoring method for early warning of mountainous roads and bridges according to claim 5 is characterized in that the laser radar sensor of the drone scans the surface deformation area to generate a deformation displacement vector map, including: The UAV continuously emits laser pulses during flight, receives reflected signals and calculates the round-trip time difference of the pulses to generate an initial point cloud data set; De-noising the initial point cloud data to remove abnormal points caused by vegetation occlusion or meteorological interference, thereby generating de-noised point cloud data; Based on the terrain slope matrix, the denoised point cloud data is subjected to elevation correction to obtain current point cloud data, thereby eliminating the error effect of terrain undulation on deformation calculation; A time series difference algorithm is used to compare the current point cloud data with the historical benchmark data to calculate the three-dimensional displacement and displacement direction of each monitoring point; The area where the displacement exceeds the warning threshold is marked as a risk area, and the risk level of the risk area is distinguished by color gradient in the deformation displacement vector diagram. 7. The inspection data monitoring method for mountain road and bridge early warning according to claim 1 is characterized in that the multi-dimensional monitoring data and the historical disaster records are input into the risk early warning model to generate a risk level signal for the target mountain road section and corresponding disposal suggestion information, including: Calling the risk warning model, matching the displacement rate in the surface deformation data with the landslide events in the historical disaster records, and triggering a first-level warning signal if the displacement rate continues to increase and the direction is consistent with the historical landslide direction; And, analyzing the trend characteristics of the road bridge crack image, if the crack extension direction coincides with the main stress direction of the bridge and the width expansion rate exceeds a critical value, a secondary warning signal is triggered; And, according to the sliding direction and speed of the slope displacement trajectory, combined with the real-time rainfall intensity data, the slope stability coefficient is calculated. If the slope stability coefficient is lower than the safety threshold, a third-level warning signal is triggered; Generate differentiated disposal suggestion information based on the warning signal level, including emergency road closure, speed limit or initiation of support structure reinforcement plan; The risk level signal and treatment suggestion information are pushed to the road and bridge management terminal, and the corresponding monitoring data traceability link is associated for manual review. 8. The inspection data monitoring method for early warning of mountainous roads and bridges according to claim 7 is characterized in that the slope stability coefficient is calculated based on the sliding direction and speed of the slope displacement trajectory in combination with real-time rainfall intensity data, including: Performing vector decomposition on the sliding direction of the slope displacement trajectory to generate time series data of horizontal displacement components and vertical displacement components; Performing similarity matching between the horizontal displacement component and the displacement pattern of landslide events corresponding to the geographic coordinates in the historical disaster records, and screening out a reference event set whose displacement trend similarity exceeds a matching threshold; Extracting the rainfall intensity time series data before each landslide event in the reference event set, and establishing a correlation mapping table between rainfall accumulation and displacement acceleration interval; Calculate the current rainfall accumulation according to the real-time rainfall intensity data, and search the association mapping table for a critical rainfall threshold value overlapping with the current displacement acceleration interval; generating a dynamic correction factor for the slope stability coefficient based on a ratio of the rate of change of the vertical displacement component to a critical rainfall threshold; The dynamic correction factor is weightedly fused with the normalized value of the slope support structure integrity parameter to output a comprehensive slope stability coefficient. 9. The inspection data monitoring method for early warning of mountainous roads and bridges according to claim 1, characterized in that the method further comprises: Establish a closed-loop feedback mechanism for the multi-dimensional monitoring data and disposal suggestion information; When the disposal suggestion information is adopted and executed by the road and bridge management terminal, the structural response data of the corresponding area is continuously monitored, and the structural response data is compared and analyzed with the expected disposal effect. If the comparative analysis result indicates that the actual improvement rate is lower than the set ratio of the expected value, the disposal plan optimization instruction is triggered; After adjusting the parameter weights of the risk warning model based on the disposal plan optimization instructions and updating the emergency avoidance node configuration in the path planning model, the optimized model parameters and configuration information are synchronized to the edge computing units of all in-service drones to realize online iterative upgrades of the monitoring strategy. 10. A server, characterized in that the server includes a processor and a memory, the memory is connected to the processor, the memory is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the memory to implement the inspection data monitoring method for mountain road and bridge early warning as described in any one of claims 1 to 9.