CN114375406A - Target speed estimation method, device and storage medium - Google Patents
Target speed estimation method, device and storage medium Download PDFInfo
- Publication number
- CN114375406A CN114375406A CN202180005033.3A CN202180005033A CN114375406A CN 114375406 A CN114375406 A CN 114375406A CN 202180005033 A CN202180005033 A CN 202180005033A CN 114375406 A CN114375406 A CN 114375406A
- Authority
- CN
- China
- Prior art keywords
- target
- speed
- boundary
- point
- vehicle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/02—Systems using the reflection of electromagnetic waves other than radio waves
- G01S17/50—Systems of measurement based on relative movement of target
- G01S17/58—Velocity or trajectory determination systems; Sense-of-movement determination systems
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
- G01S13/50—Systems of measurement based on relative movement of target
- G01S13/58—Velocity or trajectory determination systems; Sense-of-movement determination systems
- G01S13/588—Velocity or trajectory determination systems; Sense-of-movement determination systems deriving the velocity value from the range measurement
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Electromagnetism (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Physics & Mathematics (AREA)
- Optical Radar Systems And Details Thereof (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a target speed estimation method, a target speed estimation device, a control device, a computer readable storage medium, a computer program product containing instructions and a vehicle. The target speed estimation method comprises the steps of calculating a first to-be-selected speed and a second to-be-selected speed according to a first speed calculation mode and a second speed calculation mode respectively; when the target is determined to be located in a region outside a first boundary with the self-vehicle as a symmetric center, determining a first to-be-selected speed obtained according to a first speed calculation mode as the speed of the target; and otherwise, determining the second candidate speed obtained according to the second speed calculation mode as the target speed. The invention adopts two speed estimation modes to estimate the speed of the obstacle target, selects a speed estimation result as the final estimated speed of the target according to the different positions of the obstacle target relative to the vehicle, and improves the accuracy of target speed estimation.
Description
The present application claims priority from chinese patent application No. 2021103895278 entitled "target speed estimation method, apparatus, and storage medium", filed on 12/4/2021, the disclosure of which is incorporated herein by reference.
Technical Field
The present invention relates to the field of unmanned driving technology, and in particular, to a target speed estimation method, apparatus, control device, computer-readable storage medium, computer program product containing instructions, and vehicle.
Background
In the unmanned technology, there are two main ways to obtain the target speed of the obstacle at present. The first is obtained by direct measurement by a millimeter wave radar. The millimeter wave radar can accurately obtain the relative speed of the target, but the existing commonly used millimeter wave radar configuration has a certain observation blind area (usually in a range of a short distance from the left to the right of the vehicle), so that the automatic driving vehicle cannot obtain the speed information of the target in the blind area. The second method is to calculate the speed information of the target according to the position change of the obstacle target detected by the laser radar, and the method has small observation blind area, but the calculated target speed is greatly influenced by the position error measured by the laser radar.
Disclosure of Invention
The present invention is directed to a method, an apparatus, a control device, a computer-readable storage medium, a computer program product containing instructions, and a vehicle for estimating a target speed, which estimate a speed of an obstacle target using two different speed estimation methods, and select a speed estimation result as a final estimated speed of the target according to a position of the obstacle target relative to the vehicle, thereby improving accuracy of target speed estimation.
In a first aspect of the present invention, a target speed estimation method is provided, including the steps of:
calculating a first to-be-selected speed and a second to-be-selected speed according to a first speed calculation mode and a second speed calculation mode respectively;
when the target is determined to be located in a region outside a first boundary with the self-vehicle as a symmetric center, determining a first to-be-selected speed obtained according to a first speed calculation mode as the speed of the target;
when the target is determined to be located in the area within the first boundary with the self-vehicle as the symmetric center, determining a second candidate speed obtained according to the speed calculation mode II as the speed of the target;
wherein, according to the first speed calculation mode, calculating the first speed to be selected comprises:
when the target is determined to be located in a region outside a second boundary with the own vehicle as a symmetric center, the closest point of the target detected by the laser radar is used as a target tracking point; or when the target is determined to be located in an area within a second boundary with the self vehicle as a symmetric center, the target center point detected by the laser radar is used as a target tracking point;
estimating and obtaining a first speed to be selected based on the target tracking point;
and calculating a second candidate speed according to a second speed calculation mode, wherein the second candidate speed comprises the following steps:
calculating the moving speed of each target frame corner point and each target tracking point by using the position data of the target frame corner point and the target tracking point obtained by the laser radar;
and estimating a second candidate speed according to the moving speed of each target frame corner point and the target tracking point.
In a second aspect of the present invention, there is provided a target speed estimation device including:
the first speed calculation unit is used for calculating a first speed to be selected according to a first speed calculation mode;
a second speed calculating unit for calculating a second speed to be selected according to the speed calculating mode
The speed determining unit is used for determining that the first to-be-selected speed is the speed of the target when the target is located in the area outside the first boundary with the self-vehicle as the symmetry center, and otherwise, the second to-be-selected speed is determined as the speed of the target;
wherein the first speed calculation unit includes:
the tracking point determining module is used for determining that a target is located in a region outside a second boundary with the self vehicle as a symmetric center, and taking a target closest point detected by the laser radar as a target tracking point; otherwise, taking the target center point detected by the laser radar as a target tracking point;
the first speed estimation module is used for estimating and obtaining a first speed to be selected based on the target tracking point;
wherein the second speed calculation unit includes:
the target point speed calculation module is used for calculating the moving speed of each target frame corner point and each target tracking point by using the position data of the target frame corner point and the target tracking point obtained by the laser radar;
and the second speed estimation module is used for estimating a second candidate speed according to the moving speed of each target corner point and each target tracking point.
In a third aspect of the invention, there is provided a control device comprising at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the target speed estimation method of the first aspect of the invention.
In a fourth aspect of the present invention, there is provided a computer-readable storage medium containing a program or instructions for implementing the target speed estimation method according to the first aspect of the present invention when the program or instructions are run on a computer.
In a fifth aspect of the invention, there is provided a computer program product comprising instructions which, when run on a computer, cause the computer to perform the method of target speed estimation of the first aspect of the invention.
In a sixth aspect of the invention, there is provided a vehicle including the control apparatus of the third aspect of the invention.
The invention adopts two different speed estimation modes to respectively estimate the speed of the obstacle target, and then selects one speed estimation result as the final estimated speed of the target according to the different positions of the obstacle target relative to the self vehicle, thereby overcoming the problems that the automatic driving vehicle cannot obtain the speed information of the target in the millimeter wave radar blind area and the laser radar measurement result is greatly influenced by the position, and improving the accuracy of target speed estimation.
Drawings
FIG. 1 is a flow chart of a target speed estimation method according to an embodiment of the present invention;
FIG. 2 is a schematic view of a first boundary of the closed type according to the embodiment of the present invention;
FIG. 3 is a flow chart of a first candidate speed calculation according to an embodiment of the present invention;
FIG. 4 is a flowchart illustrating a second candidate speed calculation process according to an embodiment of the present invention
FIG. 5 is a schematic diagram of a first boundary arrangement according to an embodiment of the present invention;
FIG. 6 is a diagram illustrating tracking point determination and switching with a second boundary when calculating a first candidate speed according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of a velocity calculation of a uniform linear motion model according to an embodiment of the present invention;
fig. 8 is a schematic diagram of a target corner point and a target center point of the laser radar when calculating the second candidate speed according to the embodiment of the present invention;
FIG. 9 is a schematic diagram of calculating displacement velocities of a target corner point and a target center point of a laser radar and determining a target velocity according to an embodiment of the present invention;
fig. 10 is a schematic configuration diagram of a target speed estimating apparatus according to an embodiment of the present invention.
Detailed Description
The invention is described in further detail below with reference to the figures and specific examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
It should be noted that:
the "own vehicle" referred to in the present invention refers to a vehicle for executing the target speed estimation method proposed in the embodiment of the present invention, that is, a current vehicle mounted with sensor devices such as a laser radar and a millimeter wave radar.
The "target speed" referred to in the present invention is a value of a speed of a target (obstacle) relative to the vehicle in the vehicle coordinate system, and may be referred to as a relative speed of the target.
Common laser radar on the market at present, the laser that it sent does not have the penetrability, to this kind of great target of vehicle, often can only obtain the point cloud towards this one side of car certainly, can not obtain the point cloud of the one side that is shaded from the car certainly, in addition can receive the target and shelter from or the influence that shelters from each other between the object certainly, the point cloud that laser radar obtained often can't be more complete covers whole target, consequently, laser radar's target detection module usually need on the basis of the original point cloud that obtains, adopt the mode of deep learning to carry out the completion to the target point cloud, then carry out the target frame regression based on the point cloud data after the completion processing, the information such as the central point position of output target's length, width, height and target. However, when the target is farther away from the vehicle, the less the original point cloud data acquired by the laser radar is, the larger the error between the output result and the real situation is, and conversely, when the target is closer to the vehicle, the more the original point cloud data acquired by the laser radar is, the more accurate the output result is.
The conventional millimeter wave radar configuration on the automatic driving vehicle has a certain observation blind area which is usually an area with a short distance from the left to the right of the vehicle, and the speed information of a target in the blind area cannot be acquired.
In order to solve the above problem, an embodiment of the present invention provides a target speed estimation method, which calculates speeds of a target by using two different speed estimation methods based on characteristics of different sensing modules of a host vehicle, and then selects one of the calculated speeds as a target speed according to different positions of the target relative to the host vehicle.
In a first aspect of the embodiments of the present invention, a method for estimating a target speed is provided, as shown in fig. 1, the method may include the following steps:
s101: calculating a first to-be-selected speed and a second to-be-selected speed according to a first speed calculation mode and a second speed calculation mode respectively;
s102: judging the position of the target relative to the vehicle;
s103: determining the first speed to be selected or the second speed to be selected as the speed of the target according to the position of the target relative to the vehicle:
when the target is determined to be located in a region outside a first boundary with the self-vehicle as a symmetric center, determining a first to-be-selected speed obtained according to a first speed calculation mode as the speed of the target; when the target is determined to be located in the area within the first boundary with the self-vehicle as the symmetric center, determining a second candidate speed obtained according to the speed calculation mode two as the speed of the target;
in specific implementation, the "outside of the first boundary" may be set to include the first boundary itself, while the "inside of the first boundary" may be set to not include the first boundary itself; alternatively, "inside the first boundary" may be set to contain the first boundary itself, while "outside the first boundary" may be set to not contain the first boundary itself.
In specific implementation, when it is determined that the target is located in an area within a certain boundary according to basic characteristics, calibration data, test data, and the like of the laser radar, the original point cloud acquired by the laser radar is complete compared with the theoretical value (for example, the ratio of the original point cloud to the theoretical value exceeds a preset ratio threshold), and an error of the finally output result is smaller than a real situation of the target, whereas, when the target is located in an area outside the certain boundary, the original point cloud acquired by the laser radar is not complete enough compared with the theoretical value (for example, the ratio of the original point cloud to the theoretical value is lower than the preset ratio threshold), and an error of the finally output result is larger than the real situation of the target, and in this case, the boundary may be set as the first boundary in the above embodiment.
For a specific vehicle and the laser radar, the first boundary dynamically changes along with the movement of the vehicle, but the relative position of the first boundary and the vehicle is relatively stable, so that the first boundary can be determined in advance in a vehicle coordinate system according to basic characteristics, calibration data, test data and the like of the laser radar.
When the target is located in the area within the first boundary, the information such as the length, the width, the height, the central point and the like of the target output by the laser radar is more accurate, and in this case, the calculation result of the speed calculation mode II is used as the target speed; when the target is located in the area outside the first boundary, the information error of the length, width, height, center point and the like of the target output by the laser radar is large, and in this case, the calculation result of the first speed calculation mode is used as the target speed.
As an alternative embodiment, the first boundary may be a closed boundary, for example, the first boundary is a closed boundary with the origin of the own vehicle coordinate system as the symmetric center, such as a circular closed boundary, an elliptical closed boundary, or a rectangular closed boundary, or a closed boundary with other shapes, such as an elliptical boundary 300 with the origin of the own vehicle coordinate system as the symmetric center shown in fig. 2, and with respect to the position of the own vehicle 100, when the target 200 is within the elliptical boundary 300, the calculation result of the velocity calculation method two is taken as the target velocity, and when the target is located in an area outside the elliptical boundary 300, the calculation result of the velocity calculation method one is taken as the target velocity.
As an alternative, the first boundary may be a non-closed boundary, for example, the first boundary may be two linear boundaries with the origin of the vehicle coordinate system as the symmetric center, the middle area of the two linear boundaries is within the first boundary, and the areas outside the two linear boundaries are outside the first boundary.
For example, the two linear boundaries may be a first linear boundary 1 and a second linear boundary 2 which are respectively located in front of and behind the vehicle 100 and perpendicular to the driving direction of the vehicle, as shown in fig. 5, a vehicle coordinate system is established with a rear axle center of the vehicle 100 as an origin, x-axis and y-axis respectively represent a longitudinal axis and a transverse axis of the vehicle coordinate system, and for example, the first linear boundary 1 is a linear boundary located at 10m (i.e., +10m on the x-axis and parallel to the y-axis) in front of the vehicle, and the second linear boundary 2 is a linear boundary located at 10m (i.e., -10m on the x-axis and parallel to the y-axis) behind the vehicle.
In specific implementation, the first straight line boundary 1 and the second straight line boundary 2 are determined according to the complete situation of the cloud detected by the laser radar on the target. For example, according to basic characteristics, calibration data and a large amount of test data of the laser radar, when the target is located within 10m of the front or the back of the vehicle, information such as the length, the width, the height, the central point and the like of the result target output by the laser radar is accurate, and when the target is located outside 10m of the front or the back of the vehicle, the information error such as the length, the width, the height, the central point and the like of the result target output by the laser radar is large, in this case, the first straight line boundary 1 and the second straight line boundary 2 can be set to be located 10m of the front of the vehicle and 10m of the back of the vehicle respectively.
Specifically, referring to fig. 5, with respect to the host vehicle 100, the area outside the first and second straight line boundaries 1 and 2 is the first area a, the area inside the first and second straight line boundaries 1 and 2 is the second area B, and when the target 200 (denoted by id #1) is located in the first area a, the speed calculated in the first speed calculation method is used as the final target estimated speed; when the target 200 (denoted by id #1) is located in the second area B, the speed calculated by the speed calculation method two is adopted as the final target estimated speed.
The current common laser radar in the market uses a detected central point of a target as a tracking point to track the target, and is influenced by factors such as an observable position, a target clustering effect and the like, when the target is close to a self-vehicle, the laser radar can continuously, stably and accurately detect the central point of the target, and when the target is far away from the self-vehicle, the central point detected by the laser radar cannot be continuously, insufficiently stable (for example, the tracking point is lost) or inaccurate, so that the obtained target tracking result is also inaccurate.
In view of this problem, a second aspect of embodiments of the present invention provides a target tracking method for a lidar, including:
and 3, tracking the target based on the target tracking point.
Specifically, for a distant target, although the laser radar cannot continuously or accurately detect the central point of the target, the visible edge position of the target obtained by the laser radar is relatively stable, and the original point cloud data at the edge of the target can be obtained, so that the point of the target closest to the vehicle is obtained, and therefore the closest point of the target detected by the laser radar can be used as the target tracking point.
The above target tracking method for laser radar may be applied to calculate the first speed to be selected in the first speed calculation mode, as shown in fig. 3, and the calculating the first speed to be selected according to the first speed calculation mode includes:
s201: judging the position of the target relative to the vehicle;
s202: determining a target tracking point according to the position of the target relative to the self-vehicle:
when the target is determined to be located in a region outside a second boundary with the own vehicle as a symmetric center, the closest point of the target detected by the laser radar is used as a target tracking point; or when the target is determined to be located in an area within a second boundary with the self vehicle as a symmetric center, the target center point detected by the laser radar is used as a target tracking point;
s203: and estimating to obtain a first to-be-selected speed based on the target tracking point.
In specific implementation, the "outside of the second boundary" may be set to include the second boundary itself, and the "inside of the second boundary" may be set to not include the second boundary itself; alternatively, "inside the second boundary" may be set to include the second boundary itself, while "outside the second boundary" may be set to not include the second boundary itself.
In specific implementation, when it is determined that the target is located in an area within a certain boundary according to basic characteristics, calibration data, test data, and the like of the laser radar, the laser radar can continuously and accurately detect the center point of the target, whereas when the target is located in an area outside the certain boundary, the laser radar cannot continuously or accurately detect the center point of the target, in this case, the boundary may be set as the second boundary in the above embodiment.
For a specific vehicle and the laser radar, the second boundary dynamically changes along with the movement of the vehicle, but the relative position of the second boundary and the vehicle is relatively stable, so that the second boundary can be determined in advance according to basic characteristics, calibration data, test data and the like of the laser radar under a coordinate system of the vehicle.
When the target is located in the area within the second boundary, the laser radar can continuously and accurately detect the center point of the target, and the target center point detected by the laser radar is used as a target tracking point in the situation; when the target is located in the area outside the second boundary, the laser radar cannot continuously or accurately detect the central point of the target, and the closest point of the target detected by the laser radar is used as the target tracking point.
As an alternative, the second boundary may be a closed boundary, for example, the second boundary may be a closed boundary with the origin of the vehicle coordinate system as the center of symmetry, such as a circular closed boundary, an elliptical closed boundary, or a rectangular closed boundary, or other closed boundary.
As an alternative embodiment, the second boundary may also be a non-closed boundary, for example, the second boundary may also be two linear boundaries that use the origin of the vehicle coordinate system as a symmetric center, a middle area of the two linear boundaries is an area within the second boundary, and an area outside the two linear boundaries is an area outside the second boundary.
For example, the two linear boundaries corresponding to the second boundary may be a third linear boundary 3 and a fourth linear boundary 4 which are respectively located in front of and behind the vehicle 100 and perpendicular to the driving direction of the vehicle, as shown in fig. 6, a vehicle coordinate system is established with the rear axle center of the vehicle 100 as an origin, and x and y axes respectively represent the longitudinal axis and the transverse axis of the vehicle coordinate system, and illustratively, the third linear boundary 3 is a linear boundary located 2m (i.e., +2m on the x axis and parallel to the y axis) in front of the vehicle, and the fourth linear boundary 4 is a linear boundary located 2m (i.e., -2m on the x axis and parallel to the y axis) behind the vehicle.
In specific implementation, the third straight boundary 3 and the fourth straight boundary 4 are determined according to the stable situation of the laser radar setting the tracking point on the target. For example, according to basic characteristics, calibration data and a large amount of test data of the laser radar, when the target is located in an area within 2m from the front or the back of the vehicle, the laser radar can continuously and accurately detect the central point of the target, and then the target tracking result performed by taking the central point as a tracking point is also continuous and accurate, and when the target is located in an area outside 2m from the front or the back of the vehicle, the laser radar cannot continuously or accurately detect the central point of the target, and further the target tracking result performed by taking the central point as a tracking point is also not continuous and accurate, in this case, the third straight line boundary 3 and the fourth straight line boundary 4 can be set to be located 2m from the front of the vehicle and 2m from the back of the vehicle, respectively.
For the scenario shown in fig. 6, considering that the target may move around (e.g., the target vehicle suddenly decelerates, accelerates, changes lanes, etc.) near the second boundary (the third straight boundary 3, the fourth straight boundary 4), according to the aforementioned target tracking method for laser radar, the tracking point needs to be switched each time the target moves from the area outside or inside the second boundary to the area inside or outside the second boundary, however, the kalman filter needs to be reinitialized each time the tracking point is switched, which affects the tracking effect and the tracking efficiency to some extent.
In order to avoid the situation that the tracking point switches back and forth at the boundary, the embodiment of the invention provides a more optimized target tracking method for a laser radar, which comprises the following steps:
when it is determined that the target enters the region between the third straight boundary 3 and the fourth straight boundary 4 from the region outside the third straight boundary 3 located in front of the own vehicle or the region outside the fourth straight boundary 4 located behind the own vehicle, and then enters the region outside the fourth straight boundary 4 located behind the own vehicle or the region outside the third straight boundary 3 located in front of the own vehicle, the target tracking point before entering the region is still adopted when the target does not leave the region between the third straight boundary 3 and the fourth straight boundary 4, and the target tracking point is updated to the target closest point currently detected by the laser radar when the target leaves the region.
Specifically, as shown in fig. 6, the region other than the third straight boundary 3 is the first region, the region between the third straight boundary 3 and the fourth straight boundary 4 is the second region, and the region other than the fourth straight boundary 4 is the third region.
For the target 200 labeled as id # 0, when the target 200 is generated as a target (i.e., is subsequently tracked as a valid obstacle target), since the target is located in the third area, the tracking point is selected as the closest point to the target detected by the laser radar, i.e., the dot marked at the top of the rectangular block corresponding to id # 0 in the figure.
For the target 200 labeled id # 1, when the target 200 is generated as a target, since the target is located in the first area, the tracking point is also selected as the closest point of the target detected by the laser radar, i.e., the dot marked at the top of the rectangular square corresponding to id # 1 in the figure.
For the target 200 labeled id # 2, during tracking, the target enters the second region from the third region to the first region. In the third area, the target tracking point is a target closest point detected by the laser radar, that is, a dot marked at the top of a rectangular square corresponding to id # 2 in the drawing (for example, a head center point of a target vehicle); when the target passes through the fourth straight line boundary 4 and enters the second area, but does not reach the third straight line boundary 3, the target tracking point is still the closest point of the target detected by the laser radar (the dot marked at the top end of the rectangular square corresponding to id # 2 in the figure); when the target crosses the third straight line boundary 3 and enters the first area, the target tracking point is switched to a target closest point detected by the current laser radar, namely a dot marked at the bottom end of a rectangular square corresponding to id # 2 in the figure (for example, a tail center point of a target vehicle); if the target is to enter the second area again and then the third area thereafter, the tracking point is switched at the fourth straight boundary 4. By the arrangement, the situation that the target moves back and forth near the third straight line boundary 3 or the fourth straight line boundary 4 but cannot completely enter other areas can be filtered, so that the back and forth switching of tracking points caused by the situation is avoided, and the tracking efficiency and the accuracy are ensured.
As an optional embodiment, the first speed to be selected obtained by estimation based on the target tracking point in the above embodiment may be obtained by estimating, using a kalman filter, the point cloud data of the target tracking point updated in real time by the laser radar and/or the target position and speed information updated in real time by the millimeter wave radar. The step may adopt a multi-sensor information fusion technology and a target tracking technology commonly used in the art, and the embodiment of the present invention is not described again.
For the target located within the first boundary, calculating a second candidate speed according to a second speed calculation mode, as shown in fig. 4, specifically including:
s301, calculating the moving speed of each target frame corner point and each target tracking point by using the position data of the target frame corner point and the target tracking point obtained by the laser radar;
and S302, estimating a second candidate speed according to the moving speed of each target frame corner point and each target tracking point.
As shown in fig. 8, in the own vehicle coordinate system of the own vehicle 100, the target 200 (gray rectangular box) is located in the region outside the second boundary, and the target tracking point is set as the target closest point detected by the laser radar; the target frame 7 obtained by the laser radar through detection of the target 200 is a rectangular frame shown by a dotted line, and is provided with four target frame corner points 5 of an upper left side, a lower left side, an upper right side and a lower right side, and a circular point at the center of the bottom edge of the rectangular frame is a target tracking point 6. If the target 200 is located in the area within the second boundary, the target tracking point 6 is the target center point detected by the laser radar.
As shown in fig. 9, position data of each target frame corner point 5 (upper left corner point, upper right corner point, lower left corner point, lower right corner point) and position data of the target tracking point 6 obtained by the laser radar at a plurality of consecutive times (5 consecutive time stamps in fig. 9) are recorded.
As an alternative embodiment, step S301 may utilize the following uniform linear motion model to calculate the moving speed of the target corner point and the target tracking point of each timestamp according to the position data of the target corner point and the target tracking point of a plurality of consecutive timestamps:
wherein, N is an integer greater than or equal to 2 and represents the Nth time stamp;
vx is the X-axis velocity component of the angular point \ target tracking point of the target frame; vy is the Y-axis velocity component of the target frame corner point \ target tracking point; xNThe X-axis position data of the target frame angular point \ target tracking point at the Nth timestamp; y isNY-axis position data of the target frame corner point \ target tracking point at the Nth timestamp; t isNIs the time of the nth timestamp;
X0acquiring X-axis position data of a target frame corner point \ a target tracking point for a first timestamp in N continuous timestamps; y is0Y-axis position data of a target frame corner point \ a target tracking point collected for the first timestamp in the continuous N timestamps; t is0First time stamp being N consecutive time stampsThe corresponding time.
In the uniform linear motion model, at least data collected by 3 continuous time stamps (N is an integer greater than or equal to 2) are selected for calculation, so that adverse effects of the same system error possibly existing in the data collected by adjacent time stamps can be filtered, and a more accurate result is obtained.
In specific implementation, the target frame 7 obtained by the laser radar may also be a non-rectangular frame, and its corner points may not be limited to the four corner points. Optionally, several target corner points may be arbitrarily selected from all the target corner points for calculating the second candidate velocity.
As an alternative embodiment, in step S302, at least one of the moving speeds of each target corner point and the target tracking point may be determined as a second candidate speed, for example, the moving speed that can best reflect the current speed of the target among the moving speeds of the four corner points and the target tracking point is determined as the second candidate speed.
Optionally, in step S302, a second candidate speed closest to the previous timestamp in the moving speeds of the target corner point of the current timestamp and the target tracking point is determined as the second candidate speed of the current timestamp; and the second candidate speed of the initial timestamp is the speed of the target when the target is newly established. The step adopts an iterative calculation mode, and a second candidate speed which is closest to a previous timestamp in the moving speeds of a target frame corner point and a target tracking point of a current timestamp is taken as a second candidate speed of the current timestamp, so that speed estimation deviation caused by jumping of the target frame corner point detected by the laser radar can be filtered, and a smoother speed estimation value can be obtained.
As shown in fig. 9, after position data of a target frame corner point and a target tracking point obtained by the laser radar at 5 consecutive times are stored, the moving speed (including the speed component of the X axis and the speed component of the Y axis) of each target frame corner point and the target tracking point at T-1 time is calculated by using a uniform linear motion model:
moving speed of the target tracking point: vxrel _0, vyrel _0
Moving speed of upper left corner: vxrel _1, vyrel _1
Moving speed of the upper right corner point: vxrel _2, vyrel _2
Moving speed of lower left corner: vxrel _3, vyrel _3
Moving speed of lower right corner: vxrel _4, vyrel _4
After that, one group of speed is selected from the five groups of speed data corresponding to the time T-1 as the second candidate speed at the time T, for example, the second candidate speed (vxrel, vyrel) closest to the time T-1 can be used as the second candidate speed (vxrel ', vyrel') at the time T. The second candidate speed at the time T-1 is determined based on the second candidate speed at the time T-2 according to the above manner, that is, the second candidate speed at each time is determined in an iterative manner. The second candidate speed of the initial timestamp is the speed components of the X axis and the Y axis when the target is established as a new target, and may be calculated according to a uniform linear motion model, which is not described herein again.
In the embodiment of the present invention, the newly establishing a target refers to establishing a new target by using the obtained measurement information when it is determined that a virtual obstacle target meets a new target generation rule requirement (for example, a distance in a lateral direction or a longitudinal direction of the virtual obstacle target reaches a safety threshold) according to a plurality of obtained associated sensor measurement values (for example, measurement data of a laser radar, a millimeter wave radar, or a visual sensor) for the virtual obstacle target.
The embodiment of the invention adopts a target speed calculation mode of a tracking point determined based on a laser radar and a target speed calculation mode of displacement information of a frame angular point and a central point of a measured target of the laser radar to respectively estimate the speed of the obstacle target, then selects a speed estimation result as the final estimated speed of the target according to different positions of the obstacle target relative to the vehicle, can ensure higher accuracy of the calculated target speed by selecting the output target speed, and solves the technical problem that the estimation is inaccurate due to the fact that the laser radar has larger influence on the target speed estimation by the position of the target. In addition, by utilizing the characteristic that the laser radar scans the 360-degree full coverage around the vehicle, the embodiment of the invention solves the problem of blind area coverage of the millimeter wave radar from the aspect of hardware configuration.
Optionally, in the embodiment of the present invention, different strategies are adopted for estimating the speed of the newly established target according to the measurement types of the associated sensors, for example, when the measurement values associated and matched with the virtual target have the laser radar measurement, the speed of the newly established target is estimated by using the displacement and the time change of the virtual target, and when the measurement values associated and matched with the virtual target have the millimeter-wave radar measurement, the speed measurement value of the millimeter-wave radar may also be used as the speed estimation value of the newly established target.
Referring to fig. 10, a second aspect of the embodiments of the present invention provides a target speed estimation apparatus, including:
the first speed calculation unit is used for calculating a first speed to be selected according to a first speed calculation mode;
the second speed calculating unit is used for calculating a second to-be-selected speed according to a second speed calculating mode;
the speed determining unit is used for determining that the first to-be-selected speed is the speed of the target when the target is located in the area outside the first boundary with the self-vehicle as the symmetry center, and otherwise, the second to-be-selected speed is determined as the speed of the target;
wherein the first speed calculation unit includes:
the tracking point determining module is used for determining that a target is located in a region outside a second boundary with the self vehicle as a symmetric center, and taking a closest point of the target detected by the laser radar as a target tracking point; otherwise, the target center point detected by the laser radar is used as a target tracking point;
the first speed estimation module is used for estimating and obtaining a first speed to be selected based on the target tracking point;
wherein the second speed calculation unit includes:
the target point speed calculation module is used for calculating the moving speed of each target frame angular point and each target tracking point by using the displacement information of the target frame angular point and the target tracking point obtained by the laser radar;
and the second speed estimation module is used for estimating a second candidate speed according to the moving speed of each target corner point and each target tracking point.
The target speed estimation apparatus of the embodiment of the present invention may specifically adopt the target speed estimation technique disclosed in the target speed estimation method of the first aspect of the embodiment of the present invention to estimate the target speed of the obstacle, and for details, please refer to the content of the target speed estimation method disclosed in the target speed estimation method of the first aspect of the embodiment of the present invention.
In a third aspect of the embodiments of the present invention, there is provided a control device, including at least one processor; and a memory communicatively coupled to the at least one processor; the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can execute the target speed estimation method according to the first aspect of the embodiment of the present invention to estimate the target speed of the obstacle, which is not described again.
In a fourth aspect of the embodiments of the present invention, a computer-readable storage medium is provided, where the computer-readable storage medium includes a program or an instruction, and when the program or the instruction runs on a computer, the method for estimating a target speed of an obstacle according to the first aspect of the embodiments of the present invention is implemented to estimate a target speed of the obstacle, and details of the method for estimating a target speed are omitted.
In a fifth aspect of the embodiments of the present invention, a computer program product including instructions is provided, where when the computer program product runs on a computer, the computer is enabled to execute the target speed estimation method according to the first aspect of the embodiments of the present invention to perform obstacle target speed estimation, and details of the target speed estimation method are not repeated.
In a sixth aspect of the embodiments of the present invention, there is provided a vehicle including the control apparatus according to the third aspect of the embodiments of the present invention.
The invention adopts two different speed estimation modes to respectively estimate the speed of the obstacle target, and then selects a speed estimation result as the final estimated speed of the target according to the different positions of the obstacle target relative to the vehicle, thereby overcoming the problems that the automatic driving vehicle cannot obtain the speed information of the target in the millimeter wave radar blind area and the laser radar measurement result is greatly influenced by the position, and improving the precision of the target speed estimation.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (18)
1. A target speed estimation method, characterized by comprising the steps of:
calculating a first to-be-selected speed and a second to-be-selected speed according to a first speed calculation mode and a second speed calculation mode respectively;
when the target is determined to be located in a region outside a first boundary with the self-vehicle as a symmetric center, determining a first to-be-selected speed obtained according to a first speed calculation mode as the speed of the target;
when the target is determined to be located in the area within the first boundary with the self-vehicle as the symmetric center, determining a second candidate speed obtained according to the speed calculation mode II as the speed of the target;
wherein, according to the first speed calculation mode, calculating the first speed to be selected comprises:
when the target is determined to be located in a region outside a second boundary with the own vehicle as a symmetric center, the closest point of the target detected by the laser radar is used as a target tracking point; or when the target is determined to be located in an area within a second boundary with the self vehicle as a symmetric center, the target center point detected by the laser radar is used as a target tracking point;
estimating and obtaining a first speed to be selected based on the target tracking point;
and calculating a second candidate speed according to a second speed calculation mode, wherein the second candidate speed comprises the following steps:
calculating the moving speed of each target frame corner point and each target tracking point by using the position data of the target frame corner point and the target tracking point obtained by the laser radar;
and estimating a second candidate speed according to the moving speed of each target frame corner point and the target tracking point.
2. The target speed estimation method according to claim 1, wherein the first boundary is a circular closed boundary, an elliptical closed boundary, or a rectangular closed boundary having an origin of the own vehicle coordinate system as a center of symmetry.
3. The object speed estimation method according to claim 1, wherein the first boundary includes two linear boundaries centered on an origin of the own vehicle coordinate system.
4. The target speed estimation method according to claim 3, wherein the two linear boundaries include a first linear boundary and a second linear boundary that are located in front of and behind the own vehicle, respectively, and that are perpendicular to a traveling direction of the own vehicle.
5. The target speed estimation method according to claim 1, wherein the second boundary is a circular closed boundary, an elliptical closed boundary, or a rectangular closed boundary having an origin of the own vehicle coordinate system as a center of symmetry.
6. The object speed estimation method according to claim 1, wherein the second boundary includes two linear boundaries centered on an origin of the own vehicle coordinate system.
7. The target speed estimation method according to claim 6, wherein the two linear boundaries include a third linear boundary and a fourth linear boundary that are located respectively in front of and behind the own vehicle and are perpendicular to a traveling direction of the own vehicle.
8. The target speed estimation method according to claim 7, characterized in that when it is determined that the target is located in a region other than a second boundary having the own vehicle as a center of symmetry, a target closest point detected by the laser radar is taken as a target tracking point; or when determining that the target is located in an area within a second boundary with the own vehicle as a symmetric center, taking a target center point detected by the laser radar as a target tracking point, including:
determining that the target enters a region between the third straight line boundary and the fourth straight line boundary from a region outside the third straight line boundary in front of the own vehicle or from a region outside the fourth straight line boundary in back of the own vehicle, and then enters a region outside the fourth straight line boundary in back of the own vehicle or enters a region outside the third straight line boundary in front of the own vehicle; then the process of the first step is carried out,
when the target does not leave the area between the third straight line boundary and the fourth straight line boundary, still adopting a target tracking point before entering the area; and when the target leaves the area, updating the target tracking point to be the closest point of the target currently detected by the laser radar.
9. The target velocity estimation method according to claim 1, wherein estimating the first speed to be selected based on the target tracking point includes:
and estimating to obtain a first speed to be selected by utilizing the point cloud data of the target tracking point updated by the laser radar in real time and/or the target position and speed information updated by the millimeter wave radar in real time and utilizing a Kalman filter.
10. The method for estimating the target velocity according to claim 1, wherein calculating the moving velocity of each target corner point and target tracking point using the position data of the target corner point and target tracking point obtained by the lidar comprises:
and calculating the moving speed of the target frame corner point and the target tracking point of each timestamp by using a uniform linear motion model according to the position data of the target frame corner points and the target tracking points of the plurality of continuous timestamps.
11. The target speed estimation method of claim 10, wherein the consecutive plurality of time stamps are at least 3 time stamps in a row.
12. The method for estimating the target velocity according to claim 1, wherein estimating a second candidate velocity from the moving velocity of each target corner point and target tracking point comprises:
and determining at least one moving speed of the moving speeds of each target corner point and each target tracking point as a second candidate speed.
13. The method of estimating a target velocity according to claim 12, wherein determining at least one of the moving velocities of each target corner point and target tracking point as a second candidate velocity includes:
determining a second candidate speed which is closest to the previous timestamp in the moving speeds of the target frame corner point and the target tracking point of the current timestamp as the second candidate speed of the current timestamp; and the second candidate speed of the initial timestamp is the speed of the target when the target is newly established.
14. A target speed estimation device, characterized by comprising:
the first speed calculation unit is used for calculating a first speed to be selected according to a first speed calculation mode;
the second speed calculating unit is used for calculating a second to-be-selected speed according to a second speed calculating mode;
the speed determining unit is used for determining that the first to-be-selected speed is the speed of the target when the target is located in the area outside the first boundary with the self-vehicle as the symmetry center, and otherwise, the second to-be-selected speed is determined as the speed of the target;
wherein the first speed calculation unit includes:
the tracking point determining module is used for determining that a target is located in a region outside a second boundary with the self vehicle as a symmetric center, and taking a closest point of the target detected by the laser radar as a target tracking point; otherwise, the target center point detected by the laser radar is used as a target tracking point;
the first speed estimation module is used for estimating and obtaining a first speed to be selected based on the target tracking point;
wherein the second speed calculation unit includes:
the target point speed calculation module is used for calculating the moving speed of each target frame corner point and each target tracking point by using the position data of the target frame corner point and the target tracking point obtained by the laser radar;
and the second speed estimation module is used for estimating a second candidate speed according to the moving speed of each target corner point and each target tracking point.
15. A control device comprising at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the target speed estimation method of any one of claims 1-13.
16. A computer-readable storage medium, characterized by comprising a program or instructions for implementing the target speed estimation method according to any one of claims 1-13, when the program or instructions are run on a computer.
17. A computer program product comprising instructions for causing a computer to perform the target speed estimation method according to any one of claims 1-13 when the computer program product is run on the computer.
18. A vehicle characterized by comprising the control apparatus according to claim 15.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2021103895278 | 2021-04-12 | ||
| CN202110389527.8A CN115201804B (en) | 2021-04-12 | 2021-04-12 | Target speed estimation method, device and storage medium |
| PCT/CN2021/117101 WO2022217819A1 (en) | 2021-04-12 | 2021-11-11 | Target speed estimation method and apparatus, and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114375406A true CN114375406A (en) | 2022-04-19 |
| CN114375406B CN114375406B (en) | 2025-05-06 |
Family
ID=81141420
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202180005033.3A Active CN114375406B (en) | 2021-04-12 | 2021-11-11 | Target speed estimation method, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114375406B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109080628A (en) * | 2017-06-14 | 2018-12-25 | 丰田自动车株式会社 | Target determines equipment and driving assistance system |
| CN109709530A (en) * | 2017-10-26 | 2019-05-03 | 株式会社小糸制作所 | Sensor-based system and vehicle |
| US20200191942A1 (en) * | 2018-12-18 | 2020-06-18 | Hyundai Motor Company | Apparatus and method for tracking target vehicle and vehicle including the same |
-
2021
- 2021-11-11 CN CN202180005033.3A patent/CN114375406B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109080628A (en) * | 2017-06-14 | 2018-12-25 | 丰田自动车株式会社 | Target determines equipment and driving assistance system |
| CN109709530A (en) * | 2017-10-26 | 2019-05-03 | 株式会社小糸制作所 | Sensor-based system and vehicle |
| US20200191942A1 (en) * | 2018-12-18 | 2020-06-18 | Hyundai Motor Company | Apparatus and method for tracking target vehicle and vehicle including the same |
Non-Patent Citations (1)
| Title |
|---|
| 陈文强等: "基于深度神经网络的行人及骑车人联合检测", 《汽车工程》, 31 December 2018 (2018-12-31), pages 726 - 732 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114375406B (en) | 2025-05-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110705458B (en) | Boundary detection method and device | |
| US10901421B2 (en) | Method and device for detecting road boundary | |
| CN106991389B (en) | Apparatus and method for determining road edge | |
| US20200049511A1 (en) | Sensor fusion | |
| KR102547274B1 (en) | Moving robot and method for estiating location of moving robot | |
| CN104282020A (en) | Vehicle speed detection method based on target motion track | |
| KR102456626B1 (en) | Apparatus and method for traffic lane recognition in automatic steering control of vehilcles | |
| KR101558756B1 (en) | Lane Detecting Method and Apparatus for Estimating Lane Based on Camera Image and Sensor Information | |
| CN115151836A (en) | Method for detecting a moving object in the surroundings of a vehicle and motor vehicle | |
| CN114155511A (en) | Environmental information acquisition method for automatically driving automobile on public road | |
| CN110341621B (en) | Obstacle detection method and device | |
| CN111160132B (en) | Method and device for determining lane where obstacle is located, electronic equipment and storage medium | |
| Schmid et al. | Parking space detection with hierarchical dynamic occupancy grids | |
| CN113734176A (en) | Environment sensing system and method for intelligent driving vehicle, vehicle and storage medium | |
| CN114426030A (en) | Pedestrian passing intention estimation method, device and equipment and automobile | |
| TWI680898B (en) | Light reaching detection device and method for close obstacles | |
| WO2022034815A1 (en) | Vehicle surroundings recognition device | |
| CN116381632B (en) | Self-calibration method and device for radar roll angle and storage medium | |
| CN115123291B (en) | A behavior prediction method and device based on obstacle recognition | |
| JP6488226B2 (en) | Runway parameter estimation apparatus and program | |
| US20250076880A1 (en) | High-definition mapping | |
| CN118982560B (en) | Road side camera three-dimensional detection tracking method and system based on ground constraint | |
| CN114375406A (en) | Target speed estimation method, device and storage medium | |
| CN120254837A (en) | A target tracking method and device based on multi-sensor fusion | |
| Oniga et al. | A fast ransac based approach for computing the orientation of obstacles in traffic scenes |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |