CN119871456A

CN119871456A - Robot behavior control method with man-machine interaction function

Info

Publication number: CN119871456A
Application number: CN202510359110.5A
Authority: CN
Inventors: 丰卉
Original assignee: Campbell Sembcorp Beijing Technology Co ltd
Current assignee: Campbell Sembcorp Beijing Technology Co ltd
Priority date: 2025-03-25
Filing date: 2025-03-25
Publication date: 2025-04-25
Anticipated expiration: 2045-03-25
Also published as: CN119871456B

Abstract

The present invention provides a robot behavior control method with a human-computer interaction function, which relates to the field of robot technology. The robot can interact with a user, identify the user's actions through a robot terminal, find out irregular actions, and assist the user in correcting the actions through the robot. The method can be applied to various action calibration scenarios, such as dance action calibration, fitness action calibration, rehabilitation action calibration, etc. The accuracy and efficiency of training can be improved.

Description

Robot behavior control method with man-machine interaction function

Technical Field

The invention relates to the robot technology, in particular to a robot behavior control method with a man-machine interaction function.

Background

In modern life, people are increasingly concerned about the promotion of their own health and skills. Whether for rehabilitation therapy to restore body function, build good body through body building or pursue exquisite performance of dance art, accurate action calibration is a key link.

However, the conventional action calibration method has many limitations, most of which depend on one-to-one manual guidance, which not only consumes a great deal of labor and time cost, but also ensures uneven level of guidance personnel, and is difficult to ensure the accuracy and consistency of calibration.

Thus, a solution is needed to break this bottleneck.

Disclosure of Invention

The invention provides a robot behavior control method with a man-machine interaction function, which can improve the efficiency and accuracy of action calibration.

In a first aspect of the present invention, there is provided a robot behavior control method with a man-machine interaction function, including a robot body and an interaction device carrier, the robot body being provided with a plurality of electrode pads, the interaction device carrier being provided with metal parts corresponding to the electrode pads, including:

Identifying the current action of the user according to the shooting module of the robot body, and recording the non-compliance action of the user by combining the comparison result of the current action and the standard action;

Determining a calibration part corresponding to the non-compliance action, and starting an electrode plate corresponding to the calibration part to enable the electrode plate to be attached to a corresponding metal part of the interactive equipment carrier;

And controlling the robot body to perform standard actions and generating reminding information.

Optionally, in one possible implementation manner of the first aspect, the identifying, according to a shooting module of the robot body, a current action of the user, and recording, in combination with a comparison result of the current action and the standard action, an inconsistent action of the user includes:

Acquiring image data acquired by a shooting module, and extracting key points of a human body in the image data;

determining the current action of the user according to the relative position and the current angle of each key point;

and (3) calling the standard position and the standard angle of the preset point corresponding to the standard action, respectively comparing the standard position and the standard angle with the relative position and the current angle, and determining the current action with the comparison similarity smaller than the threshold value as the non-compliance action.

Optionally, in one possible implementation manner of the first aspect, before the identifying, according to the shooting module of the robot body, a current action of the user and recording, in combination with a comparison result of the current action and the standard action, an inconsistent action of the user, the method further includes:

Identifying the actual height value of the user according to the shooting module, and acquiring the proportional relationship between the hand of the user and the height;

According to the initial limb length and height of the robot body, calculating the adjustment parameters of the limb part of the robot body by combining the proportional relation and the height value, wherein the limb part comprises a hand part and a leg part;

And adjusting the limb part of the robot body according to the adjusting parameters to ensure that the height of the robot body is consistent with the height value of the user.

Optionally, in a possible implementation manner of the first aspect, in a process of identifying a current action of the user according to the shooting module of the robot body and recording an inconsistent action of the user in combination with a comparison result of the current action and the standard action, the method further includes:

Identifying key points of a human body in a multi-frame image frame acquired by a shooting module, and calculating displacement values of the same key points in adjacent frames;

calculating an average value of the plurality of displacement values, determining a preset standard displacement value corresponding to the robot body, and obtaining an adjustment coefficient according to the ratio of the average value to the standard displacement value;

And performing offset adjustment on the standard speed of the robot body according to the product of the adjustment coefficient and the standard speed of the robot body.

Acquiring a proficiency level input by a user, and controlling the action of the robot body according to a preset speed corresponding to the proficiency level;

and acquiring the current speed of the user in real time, acquiring a speed difference value between the preset speed and the current speed, and controlling the robot body to act according to the preset speed which is one level lower than the current skill level when the speed difference value is greater than the threshold value.

Optionally, in one possible implementation manner of the first aspect, determining a calibration location corresponding to the non-compliant action, and activating an electrode pad corresponding to the calibration location to attach to a corresponding metal component of the interactive apparatus carrier, including:

identifying the actual position of the electrode plate corresponding to the metal part according to the image data acquired by the shooting module;

Generating a preferred path by taking the current position of the electrode plate of the calibration part as a starting point and the actual position of the metal part as an end point;

and controlling the movement of the calibration part of the robot body according to the preferred path, starting the corresponding electrode plate, and performing the attaching operation on the electrode plate and the metal part of the calibration part.

Optionally, in a possible implementation manner of the first aspect, controlling the movement of the calibration site of the robot body according to the preferred path includes:

acquiring distance values of the corresponding electrode plates and the metal parts in the calibration part in the moving process;

When the distance value is larger than the distance threshold value, controlling the movement of the calibration part according to the preset speed;

when the distance value is smaller than or equal to the distance threshold value, the preset speed is shifted according to the product of the ratio of the distance value and the distance threshold value and the preset speed, and the movement of the calibration part is controlled according to the shifted speed.

Optionally, in a possible implementation manner of the first aspect, controlling the robot body to perform a standard action and generating the reminding information includes:

The robot body is controlled to perform standard actions, and force application values of users at corresponding positions of the electrode plates are collected based on the force feedback module;

and when the force application value is not in the standard force application value interval, generating reminding information.

Optionally, in one possible implementation manner of the first aspect, in generating the alert information when the application value is not within the standard application value interval, the alert information further includes:

When the force application value is smaller than the interval minimum value, a first adjustment value corresponding to the difference value between the interval minimum value and the force application value is obtained, and the first adjustment value is added to the current value of the corresponding electrode slice;

When the force application value is larger than the interval maximum value, a second adjustment value corresponding to the difference value between the force application value and the interval maximum value is obtained, and the second adjustment value is subtracted from the current value of the corresponding electrode plate.

Optionally, in one possible implementation manner of the first aspect, an average adjustment value of the first adjustment value and/or the second adjustment value after calibration of each electrode slice is finished is obtained, and a corresponding position of the electrode slice with the average adjustment value being greater than a preset threshold value is determined to be an error-prone high-incidence position;

And calculating the average value of the historical current values of the error-prone high-frequency positions, and setting the current value of the corresponding electrode plate as the average value in the next interactive operation.

In a second aspect of the present invention, there is provided a robot behavior control system having a man-machine interaction function, including a robot body and an interaction device carrier, the robot body being provided with a plurality of electrode pads, the interaction device carrier being provided with metal members corresponding to the electrode pads, comprising:

The identification module is used for identifying the current action of the user according to the shooting module of the robot body and recording the non-compliance action of the user by combining the comparison result of the current action and the standard action;

the attaching module is used for determining a calibration part corresponding to the non-compliance action, and starting an electrode plate corresponding to the calibration part to attach the electrode plate to a corresponding metal part of the interactive equipment carrier;

and the calibration module is used for controlling the robot body to perform standard actions and generating reminding information.

The beneficial effects of the invention are as follows:

The method comprises the steps of installing cameras at a plurality of positions of a robot body, acquiring complete user action information by using an image fusion technology, comparing the action gesture detected in real time with a pre-constructed standard action library, and accurately calculating the similarity of the action gesture and the pre-constructed standard action library, so that the non-compliance action of a user is accurately identified. The accurate action recognition and recording mode provides a reliable basis for subsequent action calibration, and the pertinence and the effectiveness of the calibration are greatly improved. For example, in dance motion calibration, fine deviation of each joint of a dancer can be accurately captured, the dancer can be assisted to rapidly improve motions, dance expressive force is improved, false motions of the dancer can be timely found in a body-building scene, the risk of sports injury is reduced, motion recovery conditions of a rehabilitation patient can be recorded in detail, and powerful data support is provided for adjustment of a rehabilitation treatment scheme.

According to the user height value and the hand and height proportional relation obtained by the shooting module, the initial limb length and height of the robot body are combined, the adjustment parameters of the limb parts (hands and legs) of the robot body are accurately calculated, the accurate matching of the height of the robot body and the height of the user is realized, and the convenience and the comfort level of human-computer interaction are greatly improved. Meanwhile, the action speed of the robot body is adjusted in various modes. The method can calculate the ratio of the average value to the standard displacement value according to the displacement value of the key points of the human body in the multi-frame images acquired by the shooting module to obtain an adjustment coefficient to offset the standard speed of the robot, and can flexibly adjust the action speed of the robot according to the preset speed corresponding to the skill level input by the user and the current speed of the user acquired in real time when the speed difference value is larger than the threshold value. The personalized speed adapting mechanism enables the robot to closely follow the action rhythm of the user, and provides customized services for users with different proficiency and movement speeds.

After determining the calibration part corresponding to the non-compliance action of the user, the robot body precisely controls the corresponding electrode plate to be attached to the metal part on the interactive equipment carrier. In the laminating process, through the distance numerical value of real-time supervision electrode slice and metal part, the dynamic adjustment calibration position velocity of movement guarantees the high efficiency and the accuracy of laminating. When the user follows the robot to perform standard actions, the force feedback module acquires force application values of the user at the corresponding positions of the electrode plates in real time. When the force application value deviates from the standard force application value interval, the current of the electrode plate is automatically adjusted. When the force application is insufficient, the current is increased to enhance the adsorption force to assist the user in applying force, and when the force application is excessive, the current is reduced to reduce the user in applying force, so that the user is helped to better complete standard actions, the training effect is improved, and the training safety is ensured. In addition, the error-prone high-incidence position is determined by calculating the average adjustment value of the adjustment values after the calibration of the electrode plates is finished, and the electrode plate current value is preset in the next interactive operation according to the historical current value average value of the position, so that the calibration effect on the error-prone position is further optimized.

Drawings

Fig. 1 is a schematic view of an application scenario provided in an embodiment of the present invention;

Fig. 2 is a schematic flow chart of a robot behavior control method with man-machine interaction function according to an embodiment of the present invention;

Fig. 3 is a schematic structural diagram of a robot behavior control system with man-machine interaction function according to an embodiment of the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

Referring to fig. 1, an application scenario is schematically shown in the embodiment of the present invention. The method and the device can interact with a user through the robot, identify actions of the user through the robot terminal, find out nonstandard actions, and assist the user in calibrating the actions through the robot. The method can be suitable for various motion calibration scenes, such as dance motion calibration, body-building motion calibration, rehabilitation motion calibration and the like.

The embodiment comprises a robot body and an interaction device carrier, wherein a plurality of electrode plates are arranged on the robot body, and the interaction device carrier is provided with metal parts corresponding to the electrode plates. Wherein the interactive apparatus carrier may be a garment with metal parts that may be sewn or embedded with thin and magnetically permeable metal patches at key joint locations, such as shoulders, elbows, wrists, hips, knees, ankles, etc. The shape and the size of the metal patches are optimized, so that the metal patches can be ensured to have enough adsorption area with the electrode pads of the robot, and normal activities of users are not influenced. For example, the metal patch of the wrist is designed to be compact and flexible, and does not interfere with the rotation of the wrist. After the user wears the interactive equipment carrier, the electrode plate in the robot body is attached to the metal part in the interactive equipment carrier, and the calibration of actions of the user can be assisted by controlling the robot body.

Referring to fig. 2, a flow chart of a robot behavior control method with a man-machine interaction function according to an embodiment of the present application is shown, and an execution subject of the method shown in fig. 2 may be a software and/or hardware device. The execution body of the present application may include, but is not limited to, at least one of a user equipment, a network device, etc. The user device may include, but is not limited to, a computer, a smart phone, a Personal Digital Assistant (PDA) DIGITAL ASSISTANT, and the above-mentioned electronic device. The network device may include, but is not limited to, a single network server, a server group of multiple network servers, or a cloud of a large number of computers or network servers based on cloud computing, where cloud computing is one of distributed computing, and a super virtual computer consisting of a group of loosely coupled computers. This embodiment is not limited thereto. The method comprises the steps of S101 to S103, and specifically comprises the following steps:

S101, recognizing the current action of the user according to the shooting module of the robot body, and recording the non-compliance action of the user by combining the comparison result of the current action and the standard action.

In practical applications, the shooting module may be a camera, and a plurality of cameras may be installed at a plurality of positions on the robot body, for example, a chest, an arm, a back, a head, etc., so as to form a distributed view angle system. The cameras at different positions can shoot the user actions from different angles, and complete user action information is obtained through an image fusion technology.

When the action training of the user is carried out, the robot can display a set of actions, the user follows the training, in the process, the actions of the user can be identified, nonstandard actions are recorded, and the nonstandard actions are the actions of the user which are inconsistent with the standard actions. The standard actions can be standard actions corresponding to the current training of the user, and the robot obtains action instructions corresponding to the standard actions in various modes. Firstly, can be connected with external equipment (such as intelligent sound box, mobile phone APP, etc.), the user selects the preset action sequence (such as body-building action package, dance action combination, etc.) to send to the robot through voice or on APP. Secondly, the robot has a motion programming function, a user can program motions on an operation interface of the robot by means of gestures, touch screens and the like, and the robot records and stores the motion instructions. Thirdly, the robot can also acquire the latest action libraries from the cloud server, and the action libraries are uploaded by professional fitness coaches, dance editors and the like, and comprise actions with various types and difficulty levels. When the robot completes one action demonstration, the actions of the user can be collected and compared with the current demonstration standard actions, so that the nonstandard actions of the user are identified and recorded.

The specific implementation manner of step S101 based on the above embodiment may be:

and acquiring image data acquired by a shooting module, and extracting key points of a human body in the image data.

Specifically, the camera on the robot body can be utilized to shoot the user action from multiple angles in real time, a continuous image sequence is obtained, and the movement information of joints and limbs of a human body can be comprehensively captured. And then, learning the characteristic modes of different parts of the human body through a deep learning detection model such as OpenPose, HRNet and the like, so that key points of the human body are accurately detected. The key points are key joint points in the human body, such as key joint points of shoulders, elbows, wrists, hips, knees, ankles and the like.

And determining the current action of the user according to the relative position and the current angle of each key point.

Through the connection relation among the key points, a skeleton structure can be constructed, and through analyzing the relative positions and the current angles of the key points in the skeleton structure, the current action gesture of the human body can be determined. For example, by calculating the angle and distance between the arm joints, it can be determined whether the arm is in a straightened state or a curved state, and the degree of curvature.

In practical application, the image sequences of various standard actions and corresponding preset point coordinate information can be collected and marked in advance, and a standard action library is constructed. Comparing the motion gesture detected in real time with the motion in the standard motion library, calculating the similarity between the motion gesture and the standard motion library, setting a proper threshold value, and judging the motion with the similarity smaller than the threshold value as the nonstandard motion.

The preset points are preset joint points, when the position distances and the angles of the corresponding joint points are compared, the similarity values of the distances and the angles are calculated, and then the calculated similarity values are added and fused to obtain the fused similarity, so that the accuracy of the judgment can be improved.

When the non-compliance action is recorded, the detected human body key point coordinates and related characteristic data such as joint angles and the like can be stored in the form of text files or databases. In order to facilitate the distinction of the key points, a unique number corresponding to each key point can be configured, and the number of the key point, the coordinate value in the image, the characteristic value and the like can be saved during recording so as to facilitate subsequent inquiry and analysis.

In addition, in order to improve the synchronization of the actions during interaction, the height of the robot body can be adjusted by combining with the height of the user before step S101, so that the height of the robot body is consistent with the height of the user, and the efficiency and convenience of man-machine interaction are improved. The height of the robot body can be adjusted specifically by the following embodiments:

And identifying the actual height value of the user according to the shooting module, and acquiring the proportional relationship between the hand and the height of the user.

Specifically, when the height value of the user is obtained, the whole body front image of the user can be shot through the shooting module, the user can be ensured to stably stand on the same plane with his feet, the body is straight, and the camera is kept vertical to the user and within a proper distance range, so that a clear and complete human body image can be obtained. Then, the acquired image is processed by using a deep learning human body key point detection algorithm, such as OpenPose, so as to identify key nodes on the body of the user, such as the top of the head and the ankle. And calculating the pixel distance between the two points according to the detected coordinate information of the vertex and ankle key points. Because the imaging principle and shooting parameters of the camera are known, the pixel distance can be converted into an actual height value through a mapping relation between the pixels and the actual distance, which is established in advance.

When the proportional relation between the hand and the height of the user is obtained, specific key points of the hand, such as shoulder peak points and finger tips, can be determined. And then, measuring the pixel distance from the shoulder peak point to the fingertip in the image, and comparing the pixel distance with the calculated pixel distance of the height of the user to obtain the proportional relationship between the hand and the height of the user, so that the user can better cooperate with the robot.

According to the initial limb length and height of the robot body, the proportional relation and the height value are combined, the adjusting parameters of the limb part of the robot body are calculated, and the limb part comprises hands and legs.

The adjusting parameter is a value corresponding to the length adjusting amplitude of the limbs of the robot body, and after the height of the user and the proportion of the limbs are obtained, the value of the limb position adjustment when the robot body is adjusted to the height corresponding to the user can be calculated by combining the value.

Because of the importance of leg length in determining height, the height difference between the user and the robot body can be calculated, and the height difference is submitted to the leg length to bear the adjustment amount, i.e. the adjustment value of the leg of the robot body can be used. For the adjustment value of the hand, the initial hand length of the robot body can be obtained by subtracting the product of the ratio of the hand of the user to the height and the height value.

The hand limbs and the leg limbs of the robot body can adopt telescopic structures, and the hand and the leg length can be adjusted by controlling motor driving.

In the process of step S101, in order to implement personalized custom training for the user, the motion speed of the robot body in the training process may be adjusted, and by comparing the error between the actual motion speed and the motion speed of the user expected to follow, the motion speed of the robot body is continuously adjusted, so that the robot body can more accurately and smoothly follow the change of the motion speed of the user, and the training accuracy is improved.

In particular, in some embodiments, the action speed adjustment of the robot body may be achieved by:

and identifying key points of human bodies in the multi-frame image frames acquired by the shooting module, and calculating displacement values of the same key points in adjacent frames.

The shooting module shoots the user action in real time, identifies and tracks the user body parts in the multi-frame image frames, acquires the position information of each key point of the user body and the change condition of the key points along with time, and can extract the characteristic parameters related to the action speed, namely the displacement values of each key point in the adjacent frames.

Calculating the average value of the displacement values, determining the preset standard displacement value corresponding to the robot body, and obtaining an adjustment coefficient according to the ratio of the average value to the standard displacement value.

It will be appreciated that since there are a plurality of keypoints, in determining the speed of the user, the determination may be made by averaging the values of the plurality of keypoints. In practical application, a standard displacement value corresponding to the standard speed during the action can be configured for the robot body in advance, an adjustment coefficient for performing offset can be obtained through the ratio of the average value to the standard displacement value, and the standard speed of the robot body can be subjected to offset adjustment through the adjustment coefficient.

By shifting the standard speed, the speed of the robot body can be adjusted according to the difference between the user action speed and the current action speed of the robot, so that the speed of the robot body is consistent with the user speed. For example, if the user's movement speed is increased, the movement speed of the robot may be increased accordingly, and if the user's movement speed is decreased, the movement speed of the robot may be decreased.

In other embodiments, the action speed adjustment of the robot body may also be achieved by:

And acquiring the proficiency level input by the user, and controlling the action of the robot body according to the preset speed corresponding to the proficiency level.

In practice, the user may input his or her skill level in a particular activity (e.g., dance, fitness, rehabilitation, etc.). The proficiency level may be a level of subjective assessment of the user's ability to perform the activity, such as by being categorized as primary, intermediate, high, etc., or may be a level determined by previous test or training records.

The corresponding speed is preset for each skill level. For example, the preset speed corresponding to the primary skill level is slower, suitable for a beginner to gradually become familiar with movements and rhythms, and the preset speed corresponding to the advanced skill level is faster, suitable for users with certain foundations and capabilities to perform more challenging exercises.

When the current speed of the user is acquired, the current speed can be acquired through an image frame acquired by the shooting module, and particularly, the current speed can be determined through the displacement value of the key point. After determining the current speed, the current speed can be compared with a preset speed corresponding to the current skill level of the user, and a speed difference between the current speed and the preset speed is calculated. This difference reflects the difference between the user's actual movement speed and the system's expected speed.

A threshold value for the speed difference may also be set. This threshold is determined based on the nature and requirements of the activity and is used to determine if the difference between the user's current speed and the preset speed is excessive. When the calculated speed difference is greater than the threshold, it is indicated that the user may not be able to keep pace with the preset speed corresponding to the current skill level, and there may be movement difficulties or other problems. In this case, therefore, the speed of the robot body may be adjusted to a preset speed one level lower than the current skill level of the user. The purpose of doing so is to make the speed of robot more accord with user's actual ability, helps the user to accomplish the activity better, avoids the user injured or can't continue situations such as training because of the speed is too fast. For example, if the user is currently of a medium skill level, when the speed difference is greater than the threshold, the robot will act at a preset speed corresponding to the primary skill level, providing a more appropriate training rhythm for the user.

S102, determining a calibration part corresponding to the non-compliance action, and starting an electrode plate corresponding to the calibration part to enable the electrode plate to be attached to a corresponding metal part of the interactive equipment carrier.

The calibration part is a body part corresponding to the irregular movement, such as an arm part. It can be understood that after determining the irregular body part of the user, after the user wears the interactive device carrier, the electrode plate of the robot body can be attached to the metal part of the corresponding part in the interactive device carrier, so that the corresponding part in the robot body can be controlled to move according to the action instruction, and the user is driven to act, and the calibration of the user action is realized.

It should be noted that, the electrode plate on the robot body is not magnetic under the condition of no power-on, so that the electrode plate on the corresponding position is powered on and started after the calibration position is determined, so that the robot body can apply force to the specific position more intensively, and accurate control and calibration of the action are realized. For example, when the hand is in fine motion, only the electrode plates near the wrist and arm joints are electrified, and the robot can accurately adjust the strength and the motion track of the parts, so that unnecessary electrification interference of other parts is avoided, and the accuracy of motion execution is improved.

Specifically, after the calibration part is determined, the position of the metal patch corresponding to the calibration part in the interactive equipment carrier can be determined, the robot body adsorbs and bonds the corresponding electrode plate and the metal part after adjusting the position of the robot body, and in the bonding process, the selected electrode plate can be electrified to generate magnetism, and the starting process is the electrifying process.

The specific implementation manner of step S102 based on the above embodiment may be:

and identifying the actual position of the electrode plate corresponding to the metal part according to the image data acquired by the shooting module.

The position of the metal part on the clothes of the user can be identified through the camera, for example, the specific position of the metal part in the image can be accurately determined through extracting the characteristics (such as shape, color, texture and the like) of the metal part in the image. And then determining the actual position of the metal component corresponding to the corresponding electrode plate, and converting the coordinates of the central point of the metal component through the mapping relation between the pixels and the actual distance to obtain the corresponding actual position.

And generating a preferable path by taking the current position of the electrode plate of the calibration part as a starting point and the actual position of the metal part as an end point.

The current position of the electrode sheet on the calibration part (for example, the body part where the electrode sheet is attached such as the arm and the leg of the user) is definitely used as the starting point of the path. The position can be obtained by the robot body through sensors such as an encoder and a gyroscope, the actual position of the metal part is taken as an end point of the path, and a path from the current position of the electrode plate to the actual position of the metal part is generated through a path planning algorithm (such as Dijkstra algorithm) based on the position information of the start point and the end point. The preferred path is the optimal path from the current location to the actual location, which may be the shortest path.

In the course of generating the path, various factors may also be considered, for example, surrounding obstacles may be collected and avoided when planning the path.

And sending an instruction to a control system of the robot body according to the generated preferred path, and controlling the robot calibration part (such as a mechanical arm and the like) to move according to the path. The drive system of the robot will precisely adjust the angle and motion of each joint according to the instructions to bring the calibration site closer to the metal part along the preferred path. During movement of the robot calibration site, electrode pads associated with the calibration site may be activated. The electrode pads may be energized to create magnetism to interact with the metal components. When the alignment site moves to the vicinity of the metal member, the electrode tab will attract each other to achieve a fit with the metal member due to the physical properties (e.g., magnetic properties) of the electrode tab. The purpose of this fitting operation is to ensure that the electrode pads are in intimate connection with the metal parts so as to be able to work properly in subsequent activities, enabling effective interaction and cooperation between the robot and the user.

In some embodiments, the calibration site movement of the robot body may be controlled according to a preferred path by:

And acquiring the distance value between the corresponding electrode plate and the metal part in the calibration part in the moving process. And when the distance value is greater than the distance threshold value, controlling the movement of the calibration part according to the preset speed. When the distance value is smaller than or equal to the distance threshold value, the preset speed is shifted according to the product of the ratio of the distance value and the distance threshold value and the preset speed, and the movement of the calibration part is controlled according to the shifted speed.

The robot body controls the movement of the calibration part according to the optimal path so as to monitor the distance between the electrode plate and the metal part continuously in the process of attaching the electrode plate and the metal part. In this process, the calibration site of the robot is constantly changing position, and the relative positions of the electrode sheet and the metal member are also dynamically changing.

When the obtained distance value is larger than the distance threshold value, the electrode plate and the metal part are relatively far, and the movement of the calibration part can be controlled according to the preset speed. The preset speed can be a relatively high and safe speed, and the purpose is to enable the robot body to quickly approach the metal part, improve the efficiency of the attaching operation, and avoid the problems of collision and the like caused by too high speed. The distance threshold may be determined according to characteristics of the electrode sheet and the metal member, fitting requirements, and movement performance of the robot.

When it is detected that the distance value between the electrode plate and the metal part is smaller than or equal to the distance threshold value, it means that the two have been relatively close. In this case, in order to more precisely control the bonding process of the electrode sheet and the metal member, it is necessary to adjust the movement speed of the calibration part in order to avoid the impact or the inaccurate bonding caused by the excessively high speed.

The ratio of the distance value to the distance threshold is calculated and then multiplied by the predetermined speed to obtain a speed offset. The principle of this calculation mode is that as the electrode plate and the metal part get closer (the distance value gets smaller), the speed offset will decrease accordingly, so that the movement speed of the calibration part gradually decreases. For example, if the distance value is half the distance threshold, the speed offset is half the preset speed, i.e., the movement speed of the calibration part becomes half the original preset speed.

And adjusting the preset speed according to the calculated speed offset to obtain the offset speed. Then, the calibration part is controlled to move continuously according to the speed after the offset, so that the electrode plate can approach the metal part at a slower and more accurate speed, and the accurate and stable attaching operation is finally realized.

Through the mode, the moving speed of the calibration part can be dynamically adjusted according to the distance value between the electrode plate and the metal part, so that the attaching efficiency is ensured, and meanwhile, the attaching precision and the attaching safety are improved.

S103, controlling the robot body to perform standard actions and generating reminding information.

After the attaching operation is completed, the robot body can control the corresponding joints to move according to the action instructions corresponding to the standard actions, so as to drive the user to act. For example, when the "lift hand to horizontal position" motion is performed, the shoulder motor and the elbow motor of the robot work cooperatively to smoothly raise the user's arm according to a preset motion trajectory and speed.

In the process of calibrating the action of the user, the robot body can continuously monitor the action state of the user through various sensors and generate reminding information when the action state deviates from a standard state, so that the user is calibrated and reminded.

In practical applications, the reminding information may be voice reminding information. For example, the robot body may have a built-in speech synthesis module, and when it is recognized that the user is not in place, a pre-recorded or real-time synthesized prompt speech is played through a speaker. For example, "your arms are not extended enough, please straighten out a little more" etc., clearly and clearly inform the user where improvement is needed.

Through the mode, the accurate calibration of actions can be realized through the interaction between the robot and the user.

The specific implementation manner of step S103 based on the above embodiment may be:

And controlling the robot body to perform standard actions, collecting force application values of users at positions corresponding to the electrode plates based on the force feedback module, and generating reminding information when the force application values are not in a standard force application value interval.

The force feedback module may be a force sensor, and the force value refers to a force value applied by a user. The force sensor can be arranged at the contact part of the robot body and the user to sense the strength and direction of the user in the following action process. For example, when the robot drives the user to perform arm lifting action, if the force applied by the user is obviously too large or too small, the force is far different from the force required by the standard action, which may indicate that the force application mode of the user is incorrect or muscle control is uncoordinated, so that the action is not standard, and at the moment, reminding information can be generated to remind the user.

The standard force application value interval may be a value interval corresponding to a normal force application set in advance, for example, when a certain joint performs rehabilitation training, a specific force application value interval is set according to a normal movement range and a muscle strength requirement of the joint so as to ensure that a user performs training within a correct strength range.

In addition, when reminding, the robot body can also assist the user in correcting action by adjusting the motion parameters of the joints of the robot body according to the difference degree and applying external force appropriately. The method specifically comprises the following examples:

When the force application value is smaller than the interval minimum value, a first adjustment value corresponding to the difference value between the interval minimum value and the force application value is obtained, the first adjustment value is added to the current value of the corresponding electrode plate, and when the force application value is larger than the interval maximum value, a second adjustment value corresponding to the difference value between the force application value and the interval maximum value is obtained, and the second adjustment value is subtracted from the current value of the corresponding electrode plate.

The force application value of each position of the user collected by the force feedback module can be continuously monitored and compared with a standard force application value interval. When the force application value at a certain position is found to be smaller than the minimum value of the standard force application value interval, the fact that the force applied by the user at the position is insufficient is indicated, and the force level required by standard action is not achieved. In order to adjust the force application condition of the user, the difference between the interval minimum value and the current force application value can be calculated. Then, according to a preset corresponding relation (for example, a mapping table of a difference value and an adjustment value established by experimental data), a first adjustment value corresponding to the difference value is obtained. This first adjustment value represents the amplitude of the current of the corresponding electrode pad that needs to be adjusted in order to bring the value of the applied force within the standard interval. After the first adjustment value is obtained, the current value of the corresponding electrode plate can be operated, and the current value is added to the first adjustment value, so that the electromagnetic adsorption force is changed, the adsorption force of the electrode plate at the position is increased, and a user is guided to move towards the correct action direction. For example, if the arm lifting angle of the user is insufficient, the robot body increases the adsorption force of the shoulder electrode plate, and assists the user in lifting the arm to a proper position.

Similarly, when the force application value at the position corresponding to a certain electrode plate is detected to be larger than the maximum value of the standard force application value interval, the fact that the force applied by the user at the position is overlarge is indicated, and the force exceeds the reasonable range allowed by the standard action. Such excessive force may cause problems such as deformation of the motion and increased risk of injury. At this time, a difference between the force application value and the interval maximum value may be calculated, and a second adjustment value corresponding to the difference may be obtained according to a preset correspondence. This second adjustment value is to adjust the electrode pad current to reduce the force applied by the user. The second adjustment value is subtracted from the present current value of the corresponding electrode pad. The reduction of the current of the electrode plate weakens the adsorption force generated by the electrode plate, so that the force applied by a user at the position is reduced, and the force application value is returned to the standard force application value interval.

Through the mode of adjusting the current of the electrode plate according to the relation between the force application value and the standard interval, when the force application of a user is abnormal, measures can be automatically taken to intervene and adjust, so that the user is helped to finish standard actions better, the training effect is improved, the safety of the user is ensured, and adverse effects caused by improper force application are avoided.

On the basis of the above embodiment, the present solution further includes the following embodiments:

The average adjustment value of the first adjustment value and/or the second adjustment value after the calibration of each electrode plate is finished is obtained, the corresponding position of the electrode plate with the average adjustment value larger than the preset threshold value is determined to be the error-prone high-frequency position, the average value of the historical current values of the error-prone high-frequency position is calculated, and the current value of the corresponding electrode plate is set to be the average value in the next interactive operation.

After the calibration operation is completed, the first adjustment value (when the force application value is smaller than the interval minimum value) and/or the second adjustment value (when the force application value is larger than the interval maximum value) generated by the fact that the force application value of the user is not matched with the standard force application value interval in the whole calibration process of each electrode slice can be recorded. These adjustment values reflect the changes made to the electrode sheet current in order to bring the applied force to the standard.

And calculating the average value of all the first adjustment values and/or the second adjustment values of each electrode slice to obtain the average adjustment value of the electrode slice. The average adjustment value comprehensively reflects the degree and trend of abnormal force application of the corresponding position of the electrode plate in the current calibration process.

In practical application, a preset threshold value may be set in advance, and the average adjustment value of each electrode sheet may be compared with the threshold value. If the average adjustment value of a certain electrode slice is larger than a preset threshold value, the fact that the force application of the corresponding position of the electrode slice is abnormal frequently and to a large extent in the calibration is indicated, and therefore the position is determined to be an error-prone high-incidence position. For example, in rehabilitation training, if the average adjustment value of the electrode sheet at the wrist is large, this indicates that the user often deviates from the standard range in the application of force of the wrist-related motion, and the wrist position is an error-prone high-incidence position. This helps to accurately locate body parts of the user that are prone to force application problems during performance of the action. The error-prone high-incidence position is the body part where the force application problem easily occurs to the user.

For the determined error-prone high-incidence position, current value records of corresponding electrode plates, namely historical current values, of the position in the past multiple interactive operations can be queried and collected. By statistically analyzing these historical current values, their average value is calculated. These historical current values reflect the current adjustment that was used during the different calibration procedures to correct the position application problem. For example, after a plurality of rehabilitation exercises, the system records the current value of the electrode plate at the wrist in each exercise, and the average value of the current values can be calculated to obtain a current value with reference significance.

The current value of the electrode plate corresponding to the error-prone high-occurrence position can be set as the calculated historical current average value before the next interactive operation (such as a new round of rehabilitation training). The purpose of this is to optimize and adjust the electrode plate current of the error-prone high-emission position in advance based on past experience. Since past calibration operations have shown that this current mean is more effective for guiding the force applied at that location to approach the standard range. For example, when a new round of rehabilitation training is started, the current of the electrode plate at the wrist is directly set to be the historical average value calculated before, so that a user can be more quickly helped to reach the standard force application requirement on the wrist action, and the training efficiency and effect are improved.

Referring to fig. 3, a schematic structural diagram of a robot behavior control system with a man-machine interaction function according to an embodiment of the present invention includes:

including robot body and mutual equipment carrier, be provided with a plurality of electrode slices on the robot body, mutual equipment carrier is provided with the metal parts that the electrode slice corresponds, include:

The apparatus of the embodiment shown in fig. 3 may be correspondingly used to perform the steps in the embodiment of the method shown in fig. 2, and the implementation principle and technical effects are similar, and are not repeated here.

It should be noted that the above embodiments are merely for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the above embodiments, it should be understood by those skilled in the art that the technical solution described in the above embodiments may be modified or some or all of the technical features may be equivalently replaced, and these modifications or substitutions do not make the essence of the corresponding technical solution deviate from the scope of the technical solution of the embodiments of the present invention.

Claims

1. The robot behavior control method with the man-machine interaction function is characterized by comprising a robot body and an interaction device carrier, wherein a plurality of electrode plates are arranged on the robot body, and the interaction device carrier is provided with metal parts corresponding to the electrode plates, and the robot behavior control method comprises the following steps:

2. The method of claim 1, wherein the step of determining the position of the substrate comprises,

Identifying the current action of the user according to the shooting module of the robot body, and recording the non-compliance action of the user by combining the comparison result of the current action and the standard action, wherein the method comprises the following steps:

3. The method of claim 1, wherein the step of determining the position of the substrate comprises,

Before the current action of the user is identified according to the shooting module of the robot body and the non-compliance action of the user is recorded by combining the comparison result of the current action and the standard action, the method further comprises the following steps:

4. The method of claim 1, wherein the step of determining the position of the substrate comprises,

In the process of identifying the current action of the user according to the shooting module of the robot body and recording the non-compliance action of the user by combining the comparison result of the current action and the standard action, the method further comprises the following steps:

5. The method of claim 1, wherein the step of determining the position of the substrate comprises,

6. The method of claim 1, wherein the step of determining the position of the substrate comprises,

Determining a calibration part corresponding to the non-compliance action, starting an electrode plate corresponding to the calibration part to attach to a corresponding metal part of the interactive device carrier, including:

7. The method of claim 6, wherein the step of providing the first layer comprises,

Controlling the movement of the calibration part of the robot body according to the preferred path, comprising:

8. The method of claim 1, wherein the step of determining the position of the substrate comprises,

The robot body is controlled to perform standard actions and generate reminding information, and the method comprises the following steps:

9. The method of claim 8, wherein the step of determining the position of the first electrode is performed,

When the force application value is not in the standard force application value interval, the process of generating the reminding information further comprises the following steps:

10. The method as recited in claim 9, further comprising:

Acquiring an average adjustment value of the first adjustment value and/or the second adjustment value after the calibration of each electrode slice is finished, and determining that the corresponding position of the electrode slice with the average adjustment value being larger than a preset threshold value is an error-prone high-incidence position;