A method and system for controlling a biomimetic robotic arm

Abstract

The application discloses a kind of bionic mechanical arm control method and system.The bionic mechanical arm control method includes the following steps: bionic space construction;Grabbing precision training;Real object grabbing training;PID control model acquisition;Bionic test.The application is applied to the automatic control system constructed by PID control algorithm, then based on the training data of the grabbing precision of bionic mechanical arm obtained by PID control algorithm training, further obtains bionic control parameter and judges, then the corresponding data reaching reference bionic control parameter is input to sensor and carries out real object grabbing training, and the grabbing precision obtained is judged, and the running track of bionic mechanical arm is controlled by acquiring PID control model, finally, real object grabbing test is carried out and the corresponding grabbing precision error is judged, the effect of improving the efficiency of bionic mechanical arm moving track planning is achieved, and the problem of low efficiency of bionic mechanical arm moving track planning in the prior art is solved.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, and in particular to a control method and system for a bionic robotic arm. Background Technology

[0002] With the rapid development of automation technology, smart factories have emerged, and robotic arms are an important component of smart factories. Controllers adjust the movement trajectory of robotic arms to avoid obstacles and complete designated tasks, thus achieving industrial production. Currently, the maintenance and installation of high-altitude towers such as power transmission towers and communication towers are still done manually. Working at heights poses significant dangers, especially given the high voltage hazard of power towers. Many institutions both domestically and internationally have conducted research on high-altitude work robots. Existing high-altitude work robots are mostly computer-controlled, and their movements cannot fully meet the operator's expectations. As the precision requirements of industrial production continue to increase, traditional feedback-based robotic arm control modes can no longer meet the needs of industrial development, making intelligent control of robotic arms a research hotspot.

[0003] Existing technologies typically employ forward and inverse kinematics calculation methods and construct digital twin bionic space methods to output collision-free joint path points for the robotic arm when completing a task, thereby realizing the robotic arm's movement trajectory planning; there are also methods based on sampling to realize the robotic arm's movement trajectory planning, achieving high-dimensional space planning with relatively small computational load.

[0004] For example, the invention patent application with publication number CN108279620A discloses a bionic arm control device and control method based on brainwaves combined with limb movements, which includes: a brainwave sensor and a motion sensing sensor connected to the control module via a wireless network; a voice broadcast module connected to the control module via a wire; and a control module connected to the bionic arm via a wireless network.

[0005] For example, the invention patent announcement CN110413107B discloses a bionic robotic hand interactive control method based on electromyography (EMG) signal pattern recognition and particle swarm optimization. This method includes: first, establishing a pattern recognition system based on surface EMG signals for pre-learning; acquiring surface EMG signals through a multi-channel EMG instrument; performing software filtering using Butterworth filtering and comb filtering; extracting feature data for various features; performing pattern recognition based on offline feature data across all channels; and using a particle swarm optimization algorithm to optimize and search for the optimal feature combination and the best channel combination. Then, acquiring real-time surface EMG signals based on the optimal feature combination and the optimal channel combination; performing filtering and feature extraction; using the KNN algorithm for real-time gesture recognition; and using the recognized gesture results to perform real-time robust control of the bionic robotic hand with each degree of freedom.

[0006] However, in the process of implementing the inventive technical solution in the embodiments of this application, it was found that the above-mentioned technology has at least the following technical problems:

[0007] In the existing technology, the bionic robotic arm control system can enable the bionic robotic arm to reach a designated position, but the control process is time-consuming, energy-intensive, and has low control efficiency, and there is a problem of low efficiency in planning the movement trajectory of the bionic robotic arm. Summary of the Invention

[0008] This application provides a bionic robotic arm control method and system, which solves the problem of low efficiency in the planning of bionic robotic arm movement trajectories in the prior art, and improves the efficiency of bionic robotic arm movement trajectory planning.

[0009] This application provides a bionic robotic arm control method, including the following steps: S1, constructing an automatic control system, which is used to determine the next position and action of the bionic robotic arm based on environmental information and the pose information of the bionic robotic arm. The automatic control system includes a sensor and a microcontroller-based bionic space. The sensor includes a position sensor and a speed sensor. The microcontroller-based bionic space is used to process and analyze the information collected by the sensor. The bionic robotic arm is used to execute a preset task; S2, training the grasping accuracy of the bionic robotic arm based on a PID control algorithm to obtain training data and record the number of training times. Based on the obtained training data, bionic control parameters are acquired. It is then determined whether the acquired bionic control parameters reach reference bionic control parameters. If the reference bionic control parameters are reached, the information collected by the sensor is input into the microcontroller-based bionic space; otherwise, training is repeated. The grasping accuracy includes grasping speed and grasping force; S3, the corresponding values ​​of the reference bionic control parameters obtained in S2 are... After the data is input to the sensor, the bionic robotic arm is trained to grasp objects, acquiring physical training data. Based on the acquired physical training data, it is determined whether the grasping accuracy of the bionic robotic arm is equal to the corresponding reference grasping accuracy. If it is equal, S4 is executed; otherwise, S2 is returned. In S4, the unit step response function is obtained by combining the obtained bionic control parameters with the microcontroller's bionic space. A PID control model is then obtained based on the obtained unit step response function. The obtained PID control model controls the running trajectory of the bionic robotic arm. The unit step response function is used to describe the error level of the bionic robotic arm's running trajectory within different time periods. In S5, the bionic robotic arm is tested to grasp objects within a preset time period, acquiring the grasping accuracy error of the bionic robotic arm. It is determined whether the acquired grasping accuracy error is less than the corresponding preset accuracy error. If it is less than the preset accuracy error, it indicates that the grasping accuracy of the bionic robotic arm has met the preset requirements; otherwise, S1 is returned. The grasping accuracy error includes grasping speed error and grasping force error.

[0010] Furthermore, the PID control algorithm is used to control the grasping accuracy of the bionic robotic arm and obtain grasping accuracy control parameters. The obtained grasping accuracy control parameters are analyzed to obtain the corresponding bionic control parameters. The grasping accuracy control includes proportional control, integral control, and derivative control. The grasping accuracy control parameters include proportional parameters, integral parameters, and derivative parameters, which are obtained through proportional control, integral control, and derivative control, respectively. The proportional parameters and integral parameters are used in conjunction with the automatic control system to determine the next position of the bionic robotic arm and generate control signals based on the errors generated during the movement.

[0011] Furthermore, the grasping accuracy error of the bionic robotic arm is calculated using the following formula: In the formula, i is the number of training sessions for the bionic robotic arm, i = 1, 2, ..., N, and N is the total number of training sessions for the bionic robotic arm. ξ i Let ψ represent the grasping speed error during the i-th training iteration of the bionic robotic arm. i Kp represents the grasping force error during the i-th training iteration of the bionic robotic arm. i K represents the proportional parameter of the i-th training iteration of the bionic robotic arm. i Kd represents the integral parameter of the i-th training iteration of the bionic robotic arm. i Let represent the differential parameters of the i-th training iteration of the bionic robotic arm, α1 represent the reference grasping speed of the bionic robotic arm, and α2 represent the reference grasping force of the bionic robotic arm. i β represents the actual grasping speed during the i-th training iteration of the bionic robotic arm. i This represents the actual grasping force during the i-th training session of the bionic robotic arm.

[0012] Furthermore, the environmental information includes binocular camera parameters, obstacle parameters, and target object parameters; the pose information is used to describe the position, velocity, and attitude information of the bionic robotic arm, specifically including binocular camera pose information, obstacle pose information, and target object pose information.

[0013] Furthermore, the error level is used to describe the oscillation degree of the unit step response function; if the error change is equal to 0, the corresponding control parameters are obtained according to the unit step response function; if the error change is not equal to 0, return to S2.

[0014] Furthermore, the specific training method for the grasping accuracy of the bionic robotic arm is as follows: S21, initialize the PID control algorithm, obtain initial bionic control parameters, and construct a PID control model based on the initial bionic control parameters; S22, obtain the pose information of the bionic robotic arm, obtain the grasping path information of the bionic robotic arm based on the pose information of the bionic robotic arm and the PID control model, and then train the grasping accuracy of the bionic robotic arm based on the obtained grasping path information and obtain the corresponding grasping accuracy; S23, determine whether the obtained grasping accuracy is equal to the reference grasping accuracy; if not equal, return to S21; if equal, obtain the corresponding bionic control parameters.

[0015] Furthermore, the specific calculation formula for the biomimetic control parameters is as follows: In the formula, i is the number of training sessions for the bionic robotic arm, i = 1, 2, ..., N, N is the total number of training sessions for the bionic robotic arm, e is the natural constant, and θ i Let ξ represent the bionic control parameters for the i-th training iteration of the bionic robotic arm. i Let ψ represent the grasping speed error during the i-th training iteration of the bionic robotic arm. i Kd represents the grasping force error during the i-th training iteration of the bionic robotic arm. i Let represent the differential parameters of the i-th training iteration of the bionic robotic arm, α1 represent the reference grasping speed of the bionic robotic arm, and α2 represent the reference grasping force of the bionic robotic arm. i β represents the actual grasping speed during the i-th training iteration of the bionic robotic arm. i This represents the actual grasping force during the i-th training session of the bionic robotic arm.

[0016] Furthermore, the specific construction method of the PID control model is as follows: The state information is obtained by initializing the PID control algorithm; a PID bionic control function is obtained based on the PID control algorithm and the obtained state information, the PID bionic control function being used to obtain the grasping accuracy error and the corresponding grasping accuracy error change rate; a PID control model is obtained based on the obtained PID bionic control function and processed by a microcontroller; in step S23, before returning to S21, the method further includes: updating the pose information of the bionic robotic arm, the reference bionic control parameters, and the PID control model.

[0017] Furthermore, the PID biomimetic control function is calculated using the following formula: In the formula, i is the index of the number of training sessions for the bionic robotic arm, i = 1, 2, ..., N, and N is the total number of training sessions for the bionic robotic arm. i Let ξ be the training time for the i-th bionic robotic arm. i Let ψ represent the grasping speed error during the i-th training iteration of the bionic robotic arm. iKp represents the grasping force error during the i-th training iteration of the bionic robotic arm. i K represents the proportional parameter of the i-th training iteration of the bionic robotic arm. i Kd represents the integral parameter of the i-th training iteration of the bionic robotic arm. i Let αi represent the differential parameters of the i-th training iteration of the bionic robotic arm, α1 represent the reference grasping speed of the i-th training iteration, and α2 represent the reference grasping force of the i-th training iteration. i β represents the actual grasping speed during the i-th training iteration of the bionic robotic arm. i δ(t) represents the actual grasping force during the i-th training iteration of the bionic robotic arm. i ) indicates that the PID controller operates at the t-th time in the bionic robotic arm. i The biomimetic control function for time, G(t) i ) indicates that the PID controller operates at the t-th time in the bionic robotic arm. i The error between the input signal and the reference bionic control parameters within a given time period, ω(t) i ) indicates that the PID controller operates at the t-th time in the bionic robotic arm. i The rate of change of the error between the input signal and the reference bionic control parameters over a time period.

[0018] This application provides a bionic robotic arm control system, which includes: a bionic space construction module, a grasping accuracy training module, a physical object grasping training module, a PID control model acquisition module, and a bionic testing module. The bionic space construction module is used to construct an automatic control system, which determines the next position and action of the bionic robotic arm based on environmental information and the posture information of the bionic robotic arm. The automatic control system includes sensors and a microcontroller-based bionic space. The sensors include position sensors and speed sensors. The microcontroller-based bionic space is used for… The bionic robotic arm processes and analyzes the information collected by the sensor to perform a preset task. The grasping accuracy training module trains the grasping accuracy of the bionic robotic arm based on a PID control algorithm, obtains training data, records the number of training iterations, acquires bionic control parameters based on the training data, and then determines whether the acquired bionic control parameters meet reference bionic control parameters. If they do, the information collected by the sensor is input into the microcontroller's bionic space; otherwise, training is repeated. The grasping accuracy includes grasping speed and grasping force. The physical object grasping training module uses... After inputting the corresponding data that achieves the reference bionic control parameters into the sensor, the bionic robotic arm undergoes real-object grasping training to acquire real-object training data. Based on the acquired real-object training data, it is determined whether the grasping accuracy of the bionic robotic arm is equal to the corresponding reference grasping accuracy. If equal, the PID control model acquisition module is executed; otherwise, it returns to the grasping accuracy training module for reprocessing. The PID control model acquisition module is used to obtain the unit step response function based on the obtained bionic control parameters and the microcontroller's bionic space, and to acquire the PID control model based on the obtained unit step response function. A PID control model controls the trajectory of the bionic robotic arm. The unit step response function is used to describe the error level of the bionic robotic arm's trajectory within different time periods. The bionic testing module is used to perform a physical grasping test on the bionic robotic arm within a preset time period to obtain the grasping accuracy error of the bionic robotic arm. It then determines whether the obtained grasping accuracy error is less than the corresponding preset accuracy error. If it is less than the preset accuracy error, it indicates that the grasping accuracy of the bionic robotic arm has met the preset requirements. Otherwise, it returns to the bionic space construction module for reprocessing. The grasping accuracy error includes grasping speed error and grasping force error.

[0019] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0020] 1. By constructing an automatic control system, and then training the bionic robotic arm to obtain bionic control parameters based on the PID control algorithm, the system makes judgments. Next, it combines the single-chip microcomputer in the automatic control system to obtain the PID control model in the bionic space. Finally, it conducts food grasping tests to obtain the grasping accuracy error. Based on the grasping accuracy error, it makes judgments again, thereby realizing the rapid planning of the bionic robotic arm's movement trajectory, thus improving the efficiency of bionic robotic arm movement trajectory planning and effectively solving the problem of low efficiency in the existing bionic robotic arm movement trajectory planning technology.

[0021] 2. By initializing the PID control algorithm and constructing a PID control model from the acquired initial bionic control parameters, the pose information of the bionic robotic arm is obtained. Based on the pose information and the PID control model, the grasping path information of the bionic robotic arm is obtained. Then, the grasping accuracy of the bionic robotic arm is trained based on the acquired grasping path information, and the corresponding grasping accuracy is obtained. Finally, the judgment is made in combination with the reference grasping accuracy, thereby achieving accurate training of the grasping accuracy of the bionic robotic arm and reducing the grasping accuracy error of the bionic robotic arm.

[0022] 3. By initializing the PID control algorithm and obtaining state information, the PID bionic control function is obtained based on the PID control algorithm and the obtained state information to obtain the grasping accuracy error and the rate of change of grasping accuracy error. Finally, based on the obtained PID bionic control function and combined with the microcontroller, the PID control model is obtained, thereby realizing the accurate construction of the PID control model and thus realizing more accurate control of the movement trajectory of the bionic robotic arm. Attached Figure Description

[0023] Figure 1 A flowchart of a biomimetic robotic arm control method provided in this application embodiment;

[0024] Figure 2 A flowchart illustrating the training process for the grasping accuracy of the bionic robotic arm provided in this application embodiment;

[0025] Figure 3 This is a schematic diagram of a bionic robotic arm control system provided in an embodiment of this application. Detailed Implementation

[0026] This application provides a bionic robotic arm control method and system, solving the problem of low efficiency in the trajectory planning of bionic robotic arms in the prior art. It constructs an automatic control system to determine the next position and action of the bionic robotic arm based on environmental information and its pose information. Then, it trains the grasping accuracy of the bionic robotic arm using a PID control algorithm to obtain training data. Based on the obtained training data, it acquires bionic control parameters, determines whether the acquired bionic control parameters meet reference bionic control parameters, and then inputs the corresponding data that meets the reference bionic control parameters into a sensor to train the bionic robotic arm to grasp objects. Physical training data is collected, and the grasping accuracy of the bionic robotic arm is determined based on the acquired physical training data to see if it is equal to the reference grasping accuracy. Then, the unit step response function is obtained based on the obtained bionic control parameters, and a PID control model is obtained based on the obtained unit step response function to control the running trajectory of the bionic robotic arm. Finally, the bionic robotic arm is tested for grasping objects within a preset time period, and the grasping accuracy error is obtained. It is determined whether the obtained grasping accuracy error is less than the corresponding preset accuracy error. If it is less than the preset accuracy error, it indicates that the grasping accuracy of the bionic robotic arm has reached the preset requirement, thus improving the efficiency of the bionic robotic arm's movement trajectory planning.

[0027] The technical solution in this application embodiment aims to address the problem of low efficiency in planning the movement trajectory of the aforementioned bionic robotic arm. The overall approach is as follows:

[0028] By constructing an automatic control system, and then training the bionic robotic arm to obtain bionic control parameters based on the PID control algorithm, the system makes judgments. Next, it combines the PID control model with the single-chip microcomputer in the automatic control system to obtain the bionic space. Finally, it conducts physical object grasping tests to obtain the grasping accuracy error. Based on the grasping accuracy error, it makes judgments again, thereby improving the efficiency of the bionic robotic arm's movement trajectory planning.

[0029] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0030] like Figure 1The diagram shows a flowchart of a bionic robotic arm control method provided in this application embodiment. The method includes the following steps: S1, Bionic Space Construction: Constructing an automatic control system. The automatic control system is used to determine the next position and action of the bionic robotic arm based on environmental information and the pose information of the bionic robotic arm. The automatic control system includes a sensor and a microcontroller bionic space. The sensor includes a position sensor and a speed sensor. The microcontroller bionic space is used to process and analyze the information collected by the sensor. The bionic robotic arm is used to execute a preset task; S2, Grasping Accuracy Training: Training the grasping accuracy of the bionic robotic arm based on a PID control algorithm to obtain training data and record the number of training times. Obtaining bionic control parameters based on the obtained training data, and then judging whether the obtained bionic control parameters reach the reference bionic control parameters. If the reference bionic control parameters are reached, the information collected by the sensor is input into the microcontroller bionic space; otherwise, retraining is performed. The grasping accuracy includes grasping speed and grasping force; S3, Physical Object Grasping Training: The information collected by the sensor in S2 is used to train the grasping accuracy of the bionic robotic arm based on the PID control algorithm to obtain training data and record the number of training times. The bionic control parameters are obtained based on the obtained training data. The bionic control parameters are then judged to reach the reference bionic control parameters. If the reference bionic control parameters are reached, the information collected by the sensor is input into the microcontroller bionic space. Otherwise, the training is repeated. The grasping accuracy includes grasping speed and grasping force; S3, Physical Object Grasping Training: The physical object grasping accuracy is trained based on the reference bionic control parameters in S2. After the corresponding data of the control parameters are input into the sensor, the bionic robotic arm is trained to grasp objects. The system acquires physical training data and determines whether the grasping accuracy of the bionic robotic arm is equal to the corresponding reference grasping accuracy. If equal, S4 is executed; otherwise, S2 is returned. S4: PID control model acquisition: Based on the obtained bionic control parameters and the microcontroller's bionic space, the unit step response function is obtained. A PID control model is then acquired based on this unit step response function. The acquired PID control model controls the trajectory of the bionic robotic arm. The unit step response function describes the error level of the bionic robotic arm's trajectory over different time periods. S5: Bionic testing: The bionic robotic arm is tested to grasp objects within a preset time period. The grasping accuracy error is acquired, and it is determined whether the acquired grasping accuracy error is less than the corresponding preset accuracy error. If less than the preset accuracy error, the grasping accuracy of the bionic robotic arm has met the preset requirements; otherwise, S1 is returned. The grasping accuracy error includes grasping speed error and grasping force error.

[0031] In this embodiment, a microcontroller is present in the bionic space, which, combined with the unit step response function, obtains a PID control model, thereby improving the efficiency of planning the movement trajectory of the bionic robotic arm.

[0032] Furthermore, the PID control algorithm is used to control the grasping accuracy of the bionic robotic arm and obtain grasping accuracy control parameters. The obtained grasping accuracy control parameters are analyzed to obtain the corresponding bionic control parameters. The grasping accuracy control includes proportional control, integral control, and derivative control. The grasping accuracy control parameters include proportional parameters, integral parameters, and derivative parameters, which are obtained through proportional control, integral control, and derivative control, respectively. The proportional parameters and integral parameters are used in conjunction with the automatic control system to determine the next position of the bionic robotic arm and generate control signals based on the errors generated during the movement.

[0033] In this embodiment, the grasping accuracy of the bionic robotic arm can be improved by reasonably adjusting the grasping accuracy control parameters, thus realizing the accurate grasping and control of different objects by the bionic robotic arm.

[0034] Furthermore, the grasping accuracy error of the bionic robotic arm is calculated using the following formula: In the formula, i is the number of training sessions for the bionic robotic arm, i = 1, 2, ..., N, and N is the total number of training sessions for the bionic robotic arm. ξ i Let ψ represent the grasping speed error during the i-th training iteration of the bionic robotic arm. i Kp represents the grasping force error during the i-th training iteration of the bionic robotic arm. i K represents the proportional parameter of the i-th training iteration of the bionic robotic arm. i Kd represents the integral parameter of the i-th training iteration of the bionic robotic arm. i Let represent the differential parameters of the i-th training iteration of the bionic robotic arm, α1 represent the reference grasping speed of the bionic robotic arm, and α2 represent the reference grasping force of the bionic robotic arm. i β represents the actual grasping speed during the i-th training iteration of the bionic robotic arm. i This represents the actual grasping force during the i-th training session of the bionic robotic arm.

[0035] In this embodiment, the reference grasping speed of the bionic robotic arm is 0.25 m / s, and the reference grasping force is 10 N, which can be adjusted according to the actual situation. The grasping accuracy error of the bionic robotic arm includes the error between the grasping of the object and the actual required grasping position and posture. When performing grasping accuracy training, various factors are comprehensively considered according to the actual situation, thereby improving the grasping accuracy of the bionic robotic arm.

[0036] Furthermore, the environmental information includes binocular camera parameters, obstacle parameters, and target object parameters; the pose information is used to describe the position, velocity, and attitude of the bionic robotic arm, specifically including binocular camera pose information, obstacle pose information, and target object pose information.

[0037] In this embodiment, the binocular camera parameters include the position, size, and orientation of the binocular camera; the obstacle parameters include the position, shape, and orientation of the obstacle; and the target object parameters include the position, shape, color, and surface features of the target object. These parameters are crucial for object recognition, localization, and tracking. In robotic arm technology, these parameters are used for the navigation, obstacle avoidance, localization, and grasping tasks of the bionic robotic arm, enabling more comprehensive planning of the bionic robotic arm's movement trajectory.

[0038] Furthermore, the error level is used to describe the degree of oscillation of the unit step response function; if the error change is equal to 0, the corresponding control parameters are obtained based on the unit step response function; if the error change is not equal to 0, return to S2.

[0039] In this embodiment, the degree of oscillation of the unit step response function is related to the grasping accuracy control parameters of the bionic robotic arm. The degree of oscillation of the unit step response function can be controlled by adjusting the proportional, integral, and derivative parameters of the system, thereby achieving efficient control of the grasping accuracy of the bionic robotic arm.

[0040] Furthermore, such as Figure 2 The diagram shows a flowchart of the training process for the grasping accuracy of the bionic robotic arm provided in this embodiment. The specific training method for the grasping accuracy of the bionic robotic arm is as follows: S21, Initialize the PID control algorithm, obtain initial bionic control parameters, and construct a PID control model based on the initial bionic control parameters; S22, Obtain the pose information of the bionic robotic arm, obtain the grasping path information of the bionic robotic arm based on the pose information and the PID control model, and then train the grasping accuracy of the bionic robotic arm based on the obtained grasping path information and obtain the corresponding grasping accuracy; S23, Determine whether the obtained grasping accuracy is equal to the reference grasping accuracy; If not equal, return to S21; If equal, obtain the corresponding bionic control parameters.

[0041] In this embodiment, the reference grasping accuracy includes the reference grasping speed and the reference grasping force. The initialization of bionic control parameters is used to initialize the parameters of the bionic control system. The requirements of different bionic control systems vary. By adjusting and optimizing the bionic control parameters, the grasping performance and effect of the bionic robotic arm are significantly improved, and high-efficiency grasping accuracy training of the bionic robotic arm is achieved.

[0042] Furthermore, the specific calculation formulas for the biomimetic control parameters are as follows: In the formula, i is the number of training sessions for the bionic robotic arm, i = 1, 2, ..., N, N is the total number of training sessions for the bionic robotic arm, e is the natural constant, and θ i Let ξ represent the bionic control parameters for the i-th training iteration of the bionic robotic arm. i Let ψ represent the grasping speed error during the i-th training iteration of the bionic robotic arm.i Kd represents the grasping force error during the i-th training iteration of the bionic robotic arm. i Let represent the differential parameters of the i-th training iteration of the bionic robotic arm, α1 represent the reference grasping speed of the bionic robotic arm, and α2 represent the reference grasping force of the bionic robotic arm. i β represents the actual grasping speed during the i-th training iteration of the bionic robotic arm. i This represents the actual grasping force during the i-th training session of the bionic robotic arm.

[0043] In this embodiment, the bionic control parameters include perception parameters, motion parameters, and control rule parameters. By obtaining corresponding data through the number of training sessions of the bionic robotic arm, the bionic control parameters of the bionic robotic arm are reasonably selected and optimized, thereby achieving more precise control over the movement trajectory of the bionic robotic arm.

[0044] Furthermore, the specific construction method of the PID control model is as follows: The state information is obtained by initializing the PID control algorithm; based on the PID control algorithm and the obtained state information, a PID bionic control function is obtained, which is used to obtain the grasping accuracy error and the corresponding rate of change of the grasping accuracy error; based on the obtained PID bionic control function and combined with the microcontroller processing, a PID control model is obtained; in S23, before returning to S21, the following is also included: updating the pose information of the bionic robotic arm, the reference bionic control parameters, and the PID control model.

[0045] In this embodiment, the acquired state information includes the position and posture information of the bionic robotic arm. The microcontroller processes and analyzes the acquired state information, and combines it with the acquired PID bionic control function to obtain a PID control model, thus realizing a more accurate construction of the PID control model.

[0046] Furthermore, the PID biomimetic control function is calculated using the following formula: In the formula, i is the index of the number of training sessions for the bionic robotic arm, i = 1, 2, ..., N, and N is the total number of training sessions for the bionic robotic arm. i Let ξ be the time for the i-th training session of the bionic robotic arm. i Let ψ represent the grasping speed error during the i-th training iteration of the bionic robotic arm. i Kp represents the grasping force error during the i-th training iteration of the bionic robotic arm. i K represents the proportional parameter of the i-th training iteration of the bionic robotic arm. i Kd represents the integral parameter of the i-th training iteration of the bionic robotic arm. i Let αi represent the differential parameters of the i-th training iteration of the bionic robotic arm, α1 represent the reference grasping speed of the i-th training iteration, and α2 represent the reference grasping force of the i-th training iteration. iβ represents the actual grasping speed during the i-th training iteration of the bionic robotic arm. i δ(t) represents the actual grasping force during the i-th training iteration of the bionic robotic arm. i ) indicates that the PID controller operates at the t-th time in the bionic robotic arm. i The biomimetic control function for time, G(t) i ) indicates that the PID controller operates at the t-th time in the bionic robotic arm. i The error between the input signal and the reference bionic control parameters within a given time period, ω(t) i ) indicates that the PID controller operates at the t-th time in the bionic robotic arm. i The rate of change of the error between the input signal and the reference bionic control parameters over a time period.

[0047] In this embodiment, the PID bionic control function is a bionic control function based on proportional, integral, and derivative control. By reasonably adjusting and optimizing the proportional, integral, and derivative parameters, precise control of the bionic robotic arm can be achieved, thereby improving system stability and the efficiency of planning the trajectory of the bionic robotic arm.

[0048] like Figure 3The diagram shows a structural schematic of a bionic robotic arm control system provided in this application embodiment. The bionic robotic arm control system includes: a bionic space construction module, a grasping accuracy training module, a physical object grasping training module, a PID control model acquisition module, and a bionic testing module. The bionic space construction module is used to construct an automatic control system. This automatic control system determines the next position and action of the bionic robotic arm based on environmental information and the posture information of the bionic robotic arm. The automatic control system includes a sensor and a microcontroller-based bionic space. The sensor includes a position sensor and a speed sensor. The microcontroller-based bionic space is used to process and analyze the information collected by the sensor. The bionic robotic arm is used to execute a preset task. The grasping accuracy training module is used to train the grasping accuracy of the bionic robotic arm based on a PID control algorithm, obtain training data, and record the number of training iterations. It acquires bionic control parameters based on the obtained training data and then determines whether the acquired bionic control parameters meet reference bionic control parameters. If they do, the information collected by the sensor is input into the microcontroller-based bionic space; otherwise, training is repeated. The grasping accuracy includes grasping speed and grasping force. The grasping training module is used to input the corresponding data that achieves the reference bionic control parameters into the sensor to train the bionic robotic arm to grasp objects. It acquires physical training data and determines whether the grasping accuracy of the bionic robotic arm is equal to the corresponding reference grasping accuracy. If they are equal, the PID control model acquisition module is executed; otherwise, it returns to the grasping accuracy training module for reprocessing. The PID control model acquisition module is used to obtain the unit step response function based on the obtained bionic control parameters and the microcontroller's bionic space, and then obtains the PID control model based on the obtained unit step response function. The PID control model is used to control the trajectory of the bionic robotic arm. The unit step response function is used to describe the error level of the trajectory of the bionic robotic arm in different time periods. The bionic testing module is used to perform physical grasping tests on the bionic robotic arm in a preset time period to obtain the grasping accuracy error of the bionic robotic arm. It is then determined whether the obtained grasping accuracy error is less than the corresponding preset accuracy error. If it is less than the preset accuracy error, it indicates that the grasping accuracy of the bionic robotic arm has met the preset requirements. Otherwise, it returns to the bionic space construction module for reprocessing. The grasping accuracy error includes grasping speed error and grasping force error.

[0049] In this embodiment, the automatic control system refers to the automatic control of key parameters in the training of the bionic robotic arm without direct human intervention. The bionic robotic arm is a mechanism and device that mimics a biological arm, possessing a similar structure and function. The components of the bionic robotic arm include: a base; a main body assembly positioned in the center of the base, with two or more support seats vertically arranged on the main body assembly, and two or more support arms movably arranged between each pair of adjacent support seats; and a torsion assembly positioned on the base and circumferentially on the main body assembly, with each torsion assembly movably equipped with... The system comprises several sets of cycloids and a drive component equal in number to the number of cycloids, capable of controlling the tension of the cycloids. Each set of cycloids is interwoven on a support base. When the cycloid on one side of the support base tightens and the cycloid on the other side of the support base relaxes, it can cause the support base and support arm to bend towards the side of the tightened cycloid. A rotating component is mounted on the base, with a mechanical claw at one end, which is rotatably mounted on the top of the main component. The bionic robotic arm can achieve bending, extension, and grasping similar to a biological arm. The bionic robotic arm has broad application prospects, providing better services for human production and life, and realizing the intelligence and adaptability of robots.

[0050] The technical solutions described in the embodiments of this application have at least the following technical effects or advantages: Compared with the bionic arm control device and control method based on EEG combined with limb movements disclosed in the invention patent application with publication number CN108279620A, the embodiments of this application initialize the PID control algorithm and construct a PID control model from the obtained initial bionic control parameters. Then, the pose information of the bionic robotic arm is obtained, and the grasping path information of the bionic robotic arm is obtained based on the pose information and the PID control model. Then, the grasping accuracy of the bionic robotic arm is trained based on the obtained grasping path information and the corresponding grasping accuracy is obtained. Finally, the grasping accuracy is judged in combination with the reference grasping accuracy, thereby achieving accurate training of the bionic arm. The grasping accuracy of the robotic arm is improved, thereby reducing the grasping accuracy error of the bionic robotic arm. Compared with the bionic robotic hand interactive control method based on electromyographic signal pattern recognition and particle swarm optimization announced in the invention patent announcement CN110413107B, the embodiment of this application initializes the PID control algorithm and obtains the state information. Then, based on the PID control algorithm and the obtained state information, a PID bionic control function is obtained to obtain the grasping accuracy error and the rate of change of the grasping accuracy error. Finally, based on the obtained PID bionic control function and combined with the microcontroller processing, a PID control model is obtained, thereby realizing the accurate construction of the PID control model and thus achieving more accurate control of the movement trajectory of the bionic robotic arm.

[0051] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0052] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0053] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0054] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0055] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0056] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A biomimetic robotic arm control method, characterized in that, Includes the following steps: S1. Construct an automatic control system. The automatic control system is used to determine the next position and action of the bionic robotic arm based on environmental information and the pose information of the bionic robotic arm. The automatic control system includes a sensor and a microcontroller bionic space. The sensor includes a position sensor and a speed sensor. The microcontroller bionic space is used to process and analyze the information collected by the sensor. The bionic robotic arm is used to perform a preset task. S2, the grasping accuracy of the bionic robotic arm is trained based on the PID control algorithm to obtain training data and record the number of training times of the bionic robotic arm. Based on the obtained training data, the bionic control parameters are obtained, and it is then determined whether the obtained bionic control parameters reach the reference bionic control parameters. If the reference bionic control parameters are reached, the information collected by the sensor is input into the microcontroller's bionic space; otherwise, the training is repeated. The grasping accuracy includes grasping speed and grasping force. S3: Input the corresponding data of the reference bionic control parameters in S2 into the sensor and then train the bionic robotic arm to grasp objects. Obtain the physical training data and determine whether the grasping accuracy of the bionic robotic arm is equal to the corresponding reference grasping accuracy based on the obtained physical training data. If it is equal, execute S4; otherwise, return to S2. S4. Based on the obtained bionic control parameters and the single-chip microcomputer bionic space, the unit step response function is obtained, and the PID control model is obtained based on the obtained unit step response function. The running trajectory of the bionic robotic arm is controlled by the obtained PID control model. The unit step response function is used to describe the error degree of the running trajectory of the bionic robotic arm in different time periods. S5, within a preset time period, perform a physical grasping test on the bionic robotic arm, obtain the grasping accuracy error of the bionic robotic arm, and determine whether the obtained grasping accuracy error is less than the corresponding preset accuracy error. If it is less than the preset accuracy error, it indicates that the grasping accuracy of the bionic robotic arm has reached the preset requirement; otherwise, return to S1. The grasping accuracy error includes grasping speed error and grasping force error. The grasping accuracy error of the bionic robotic arm is calculated using the following formula: ， ， In the formula, This is a number representing the number of training sessions for the bionic robotic arm. , This represents the total number of training sessions for the bionic robotic arm. Indicates the first Grasping speed error during secondary training of the bionic robotic arm Indicates the first Grasping force error during secondary training of the bionic robotic arm Indicates the first The proportional parameters of the bionic robotic arm training Indicates the first Integral parameters for training the bionic robotic arm. Indicates the first Differential parameters of the bionic robotic arm training This indicates the reference grasping speed of the bionic robotic arm. This indicates the reference grasping force of the bionic robotic arm. Indicates the first The actual grasping speed of the second bionic robotic arm training. Indicates the first The actual grasping force of the bionic robotic arm during training.

2. The bionic robotic arm control method as described in claim 1, characterized in that: The PID control algorithm is used to control the grasping accuracy of the bionic robotic arm and obtain grasping accuracy control parameters. The obtained grasping accuracy control parameters are analyzed to obtain the corresponding bionic control parameters. The grasping accuracy control includes proportional control, integral control, and derivative control; The grasping accuracy control parameters include proportional parameters, integral parameters, and derivative parameters, which are obtained through proportional control, integral control, and derivative control, respectively. The proportional and integral parameters are used to generate control signals based on the errors generated when the automatic control system determines the next position and movement of the bionic robotic arm.

3. The bionic robotic arm control method as described in claim 1, characterized in that: The environmental information includes binocular camera parameters, obstacle parameters, and target object parameters; The pose information is used to describe the position, velocity, and attitude information of the bionic robotic arm, specifically including the pose information of the binocular camera, the pose information of the obstacle, and the pose information of the target object.

4. The bionic robotic arm control method as described in claim 1, characterized in that: The error level is used to describe the degree of oscillation of the unit step response function; If the error change is equal to 0, the corresponding control parameters are obtained based on the unit step response function; If the error change is not equal to 0, return to S2.

5. The bionic robotic arm control method as described in claim 2, characterized in that: The specific training method for the grasping accuracy of the bionic robotic arm is as follows: S21, Initialize the PID control algorithm, obtain initial biomimetic control parameters, and construct a PID control model from the initial biomimetic control parameters; S22, Obtain the pose information of the bionic robotic arm, obtain the grasping path information of the bionic robotic arm based on the pose information of the bionic robotic arm and the PID control model, and then train the grasping accuracy of the bionic robotic arm based on the obtained grasping path information and obtain the corresponding grasping accuracy. S23, Determine whether the obtained crawling precision is equal to the reference crawling precision; If not equal, return S21; If equal, then obtain the corresponding biomimetic control parameters.

6. The bionic robotic arm control method as described in claim 1, characterized in that, The specific calculation formula for the biomimetic control parameters is as follows: ， In the formula, This is a number representing the number of training sessions for the bionic robotic arm. , This represents the total number of training sessions for the bionic robotic arm. It is a natural constant. Indicates the first Biomimetic control parameters for the training of the biomimetic robotic arm. Indicates the first Grasping speed error during secondary training of the bionic robotic arm Indicates the first Grasping force error during secondary training of the bionic robotic arm Indicates the first Differential parameters of the bionic robotic arm training This indicates the reference grasping speed of the bionic robotic arm. This indicates the reference grasping force of the bionic robotic arm. Indicates the first The actual grasping speed of the second bionic robotic arm training. Indicates the first The actual grasping force of the bionic robotic arm during training.

7. The bionic robotic arm control method as described in claim 5, characterized in that: The specific construction method of the PID control model is as follows: The state information is obtained by initializing the PID control algorithm; Based on the PID control algorithm and combined with the acquired state information, a PID bionic control function is obtained. The PID bionic control function is used to obtain the grasping accuracy error and the corresponding grasping accuracy error change rate. Based on the obtained PID bionic control function, and combined with the processing of the single-chip microprocessor, a PID control model is obtained; S23 further includes updating the pose information of the bionic robotic arm, the reference bionic control parameters, and the PID control model before returning to S21.

8. The bionic robotic arm control method according to claim 7, characterized in that, The PID biomimetic control function is calculated using the following formula: ， In the formula, This is a number representing the number of training sessions for the bionic robotic arm. , This represents the total number of training sessions for the bionic robotic arm. For the first The training time for the bionic robotic arm. Indicates the first Grasping speed error during secondary training of the bionic robotic arm Indicates the first Grasping force error during secondary training of the bionic robotic arm Indicates the first The proportional parameters of the bionic robotic arm training Indicates the first Integral parameters for training the bionic robotic arm. Indicates the first Differential parameters of the bionic robotic arm training Indicates the first Reference grasping speed for the second-generation bionic robotic arm training Indicates the first Reference gripping force for the second bionic robotic arm training Indicates the first The actual grasping speed of the second bionic robotic arm training. Indicates the first The actual grasping force of the bionic robotic arm during training. This indicates that the PID controller is in the first stage of the bionic robotic arm. A biomimetic control function for time. This indicates that the PID controller is in the first stage of the bionic robotic arm. The error between the input signal and the reference bionic control parameters within a given time period. This indicates that the PID controller is in the first stage of the bionic robotic arm. The rate of change of the error between the input signal and the reference bionic control parameters over a time period.

9. A bionic robotic arm control system, characterized in that: The bionic robotic arm control system includes: a bionic space construction module, a grasping accuracy training module, a physical object grasping training module, a PID control model acquisition module, and a bionic testing module; The bionic space construction module is used to construct an automatic control system. The automatic control system is used to determine the next position and action of the bionic robotic arm based on environmental information and the pose information of the bionic robotic arm. The automatic control system includes a sensor and a microcontroller bionic space. The sensor includes a position sensor and a speed sensor. The microcontroller bionic space is used to process and analyze the information collected by the sensor. The bionic robotic arm is used to perform a preset task. The grasping accuracy training module is used to train the grasping accuracy of the bionic robotic arm based on the PID control algorithm, obtain training data, record the number of training times of the bionic robotic arm, obtain bionic control parameters based on the obtained training data, and then determine whether the obtained bionic control parameters reach the reference bionic control parameters. If the reference bionic control parameters are reached, the information collected by the sensor is input into the microcontroller's bionic space; otherwise, the training is repeated. The grasping accuracy includes grasping speed and grasping force. The physical grasping training module is used to input the corresponding data that reaches the reference bionic control parameters into the sensor to train the bionic robotic arm to grasp objects, acquire physical training data, and determine whether the grasping accuracy of the bionic robotic arm is equal to the corresponding reference grasping accuracy based on the acquired physical training data. If it is equal, the PID control model acquisition module function is executed; otherwise, it returns to the grasping accuracy training module for reprocessing. The PID control model acquisition module is used to obtain the unit step response function based on the obtained bionic control parameters and the microcontroller bionic space, and to obtain the PID control model based on the obtained unit step response function. The obtained PID control model is used to control the running trajectory of the bionic robotic arm. The unit step response function is used to describe the error degree of the running trajectory of the bionic robotic arm in different time periods. The bionic testing module is used to perform a physical grasping test on the bionic robotic arm within a preset time period, obtain the grasping accuracy error of the bionic robotic arm, and determine whether the obtained grasping accuracy error is less than the corresponding preset accuracy error. If it is less than the preset accuracy error, it indicates that the grasping accuracy of the bionic robotic arm has met the preset requirements; otherwise, it returns to the bionic space construction module for reprocessing. The grasping accuracy error includes grasping speed error and grasping force error. The grasping accuracy error of the bionic robotic arm is calculated using the following formula: ， ， In the formula, This is a number representing the number of training sessions for the bionic robotic arm. , This represents the total number of training sessions for the bionic robotic arm. Indicates the first Grasping speed error during secondary training of the bionic robotic arm Indicates the first Grasping force error during secondary training of the bionic robotic arm Indicates the first The proportional parameters of the bionic robotic arm training Indicates the first Integral parameters for training the bionic robotic arm. Indicates the first Differential parameters of the bionic robotic arm training This indicates the reference grasping speed of the bionic robotic arm. This indicates the reference grasping force of the bionic robotic arm. Indicates the first The actual grasping speed of the second bionic robotic arm training. Indicates the first The actual grasping force of the bionic robotic arm during training.

Images