CN-121973244-A - Self-adaptive optimization robot control method and system for hand-eye calibration

CN121973244ACN 121973244 ACN121973244 ACN 121973244ACN-121973244-A

Abstract

The application relates to the technical field of robots, and discloses a robot control method and a system for self-adaptive optimization of hand-eye calibration, wherein the method comprises the steps of obtaining a calibration instruction and operation constraint data of a target robot; the method comprises the steps of performing hand-eye calibration matrix calculation of a target robot according to a calibration instruction and operation constraint data by adopting an adaptive optimization path strategy based on online quality feedback and a preset calibration error compensation strategy to obtain a calibration result, obtaining a working instruction of the target robot, controlling the target robot to operate according to the calibration result, and compensating the calibration result according to the preset operation drift compensation strategy in the operation process of the target robot. Through the synergistic effect of the self-adaptive optimization of the acquisition process, the error compensation of the calibration result and the drift compensation of the operation stage, the robot can maintain the operation capability with high precision and stable precision under the complex working condition.

Inventors

QIU GUODONG
CHEN WEIYUN
Tai Yunxin

Assignees

跨维(深圳)智能数字科技有限公司

Dates

Publication Date: 20260505
Application Date: 20260403

Claims (10)

1. A method of robot control for adaptive optimization of hand-eye calibration, the method comprising: acquiring a calibration instruction and operation constraint data of a target robot; Performing hand-eye calibration matrix calculation of the target robot according to the calibration instruction and the operation constraint data by adopting an adaptive optimization path strategy based on online quality feedback and a preset calibration error compensation strategy to obtain a calibration result; acquiring a working instruction of the target robot; Responding to the working instruction, controlling the target robot to operate according to the calibration result, and compensating the calibration result according to a preset operation drift compensation strategy in the operation process of the target robot; The self-adaptive optimization path strategy based on online quality feedback, the calibration error compensation strategy and the operation drift compensation strategy form a cross-link closed-loop control system in a cooperative mode.
2. The method for adaptively optimizing a hand-eye calibration according to claim 1, wherein the step of calculating a hand-eye calibration matrix of the target robot according to the calibration command and the operation constraint data to obtain a calibration result by adopting an adaptive optimization path strategy based on online quality feedback and a preset calibration error compensation strategy comprises the steps of: acquiring the calibration pose of the calibration object according to the calibration instruction, and generating a sampling path sequence according to the calibration pose; Controlling the target robot to move along the sampling path sequence; When the target robot moves along the sampling path sequence to reach a sampling point, controlling a target vision device to shoot the calibration object to obtain a shooting result and a single sampling point tail end pose of the target robot, and performing quality assessment according to the shooting result to obtain an assessment result; If the conclusion of the evaluation result is qualified, taking the shooting result and the single sampling point tail end pose which are qualified in the conclusion of the evaluation result as qualified data pairs, and jumping to the step of controlling the target robot to move along the sampling path sequence to continue to execute until no sampling point which is not reached by the target robot exists in the sampling path sequence; If the conclusion of the evaluation result is that the operation constraint data, the calibration pose, the current evaluation result and the current single sampling point end pose are not qualified, updating the sampling path sequence to enable the target robot to move to a better observation pose, and jumping to the step of controlling the target robot to move along the sampling path sequence to continue to be executed until no sampling point which is not reached by the target robot exists in the sampling path sequence; Based on the calibration error compensation strategy, calculating a hand-eye calibration matrix of the target robot according to the calibration pose and each qualified data to obtain the calibration result; The target visual equipment is located at the tail end of the mechanical arm of the target robot, the calibration object is located at a preset position outside the target robot, or the target visual equipment is located at a preset position outside the target robot, and the calibration object is located at the tail end of the mechanical arm of the target robot.
3. The method for adaptively optimizing a hand-eye calibration according to claim 2, wherein the step of calculating a hand-eye calibration matrix of the target robot based on the calibration error compensation strategy according to the calibration pose and the respective qualified data to obtain the calibration result comprises: according to the target pose and each qualified data pair, performing hand-eye calibration matrix calculation of the target robot to obtain a first matrix; acquiring data from a historical calibration data set corresponding to the target robot according to a preset first time window, and taking the data as first data; inputting the first data into a pre-trained first prediction model to predict drift trend of a hand-eye calibration matrix, so as to obtain a first prediction result; Generating a calibration compensation matrix according to the first prediction result to serve as a second matrix; And according to the second matrix, performing calibration error compensation on the first matrix to obtain the calibration result.
4. The method for adaptively optimizing a hand-eye calibration of claim 3, wherein said step of calculating a hand-eye calibration matrix of said target robot based on said calibration pose and each of said pair of qualified data to obtain a first matrix, further comprises: acquiring a historical calibration data set of the target robot; and calculating a hand-eye calibration matrix of the target robot according to the historical calibration data set, the calibration pose and the qualified data pairs to obtain the first matrix.
5. The method for adaptively optimizing a robot control for hand-eye calibration according to claim 2, wherein the step of acquiring calibration instructions and job constraint data of the target robot comprises: based on a preset calibration starting condition, acquiring a calibration instruction and operation constraint data of the target robot; The calibration starting conditions comprise one or more of a calibration signal input by a user, a calibration signal triggered by adjustment of target visual equipment, a calibration signal triggered by adjustment of the target robot, a difference value between the current operation temperature corresponding to the calibration result and the calibration temperature is larger than a preset temperature value, and the current operation precision corresponding to the calibration result is larger than the preset precision and the active prevention starting conditions.
6. The method for adaptively optimizing a robot control system for hand-eye calibration according to claim 5, wherein the calibration initiation conditions include an active preventive initiation condition, and wherein the method comprises: Acquiring a prediction instruction, responding to the prediction instruction, and acquiring data from a historical calibration data set corresponding to the target robot according to a preset second time window to serve as second data; Inputting the second data into a pre-trained first prediction model to predict drift trend of a hand-eye calibration matrix to obtain a second prediction result, and determining evaluation time according to the second prediction result and a preset first drift threshold; Acquiring actual drift data of a hand-eye calibration matrix of the target robot according to the evaluation time, and judging whether the target robot is calibrated according to the actual drift data and a preset second drift threshold value to obtain a judgment result; If the judgment result is yes, acquiring a calibration instruction and operation constraint data of the target robot, and taking the second prediction result as prior information for guiding the generation of the sampling path sequence so that the sampling path sequence has higher sampling point density in the drift direction indicated by the second prediction result; and if the judgment result is negative, generating the next prediction instruction according to the actual drift data, the second drift threshold value and the second prediction result.
7. The method for adaptively optimizing a robot control for hand-eye calibration according to claim 1, wherein the step of compensating the calibration result according to a preset operation drift compensation strategy during the operation of the target robot comprises: In the process of operating the target robot, acquiring an operation drift compensation matrix as a third matrix according to a preset time interval, and compensating the calibration result according to the third matrix; the third matrix is a calibration compensation matrix generated according to the operation data corresponding to the working instruction and the equipment parameters of the target robot; the job data includes one or more of job duration, job intensity, job environment temperature, job vibration data, job height, job depth, and job load data.
8. A robot control system for adaptive optimization of hand-eye calibration, characterized in that the system comprises a system controller and a target robot, the system controller being communicatively connected to the target robot, the system controller being configured to implement the steps of the robot control method for adaptive optimization of hand-eye calibration according to any of claims 1 to 7.
9. A target robot, characterized in that the robot comprises a memory, a processor and a computer program stored in the memory and executable on the processor, which processor, when executing the computer program, realizes the steps of the robot control method for adaptive optimization of hand-eye calibration according to any of claims 1 to 7.
10. A computer-readable storage medium, characterized in that it stores a computer program which, when executed by a processor, implements the steps of the robot control method for adaptive optimization of hand-eye calibration according to any of claims 1 to 7.

Description

Self-adaptive optimization robot control method and system for hand-eye calibration Technical Field The application relates to the technical field of robots, in particular to a robot control method and system for self-adaptive optimization of hand-eye calibration. Background In the technical field of robots, hand-eye calibration is a key step for realizing precise operation of the robot, and aims to establish a coordinate transformation relationship between a vision sensor and an actuator (or a robot base coordinate system) at the tail end of the robot. The traditional hand-eye calibration method generally relies on a preset fixed path to collect data, lacks a real-time evaluation and feedback mechanism for image quality in the sampling process, and is difficult to cope with dynamic interference factors such as ambient illumination change, shielding, reflection and the like, so that the calibration precision is unstable. In addition, in the prior art, after the hand-eye calibration is finished, the hand-eye relation drift caused by factors such as mechanical abrasion, temperature change and the like in the long-term operation process of the robot is often ignored, and the continuous precision and stability in the operation process are difficult to ensure. Disclosure of Invention Based on the above, it is necessary to solve the technical problems that in the prior art, the calibration precision is unstable by a hand-eye calibration mode with a fixed sampling path, and the hand-eye relation drift of a robot in a long-term operation process is ignored, so that the accuracy of the precision and the stability of the precision in the operation process are difficult to ensure, and a self-adaptive optimization robot control method and a self-adaptive optimization system for the hand-eye calibration are provided. In a first aspect, a method for adaptively optimizing robot control for hand-eye calibration is provided, the method comprising: acquiring a calibration instruction and operation constraint data of a target robot; Performing hand-eye calibration matrix calculation of the target robot according to the calibration instruction and the operation constraint data by adopting an adaptive optimization path strategy based on online quality feedback and a preset calibration error compensation strategy to obtain a calibration result; acquiring a working instruction of the target robot; Responding to the working instruction, controlling the target robot to operate according to the calibration result, and compensating the calibration result according to a preset operation drift compensation strategy in the operation process of the target robot; The self-adaptive optimization path strategy based on online quality feedback, the calibration error compensation strategy and the operation drift compensation strategy form a cross-link closed-loop control system in a cooperative mode. In a second aspect, there is provided a robot control system for adaptive optimization of hand-eye calibration, the system comprising a system controller and a target robot, the system controller being communicatively connected to the target robot, the system controller being configured to implement the steps of the robot control method for adaptive optimization of hand-eye calibration as described above. In a third aspect, a target robot is provided, the robot comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the robot control method for adaptive optimization of hand-eye calibration as described above when executing the computer program. In a fourth aspect, a computer readable storage medium is provided, storing a computer program which, when executed by a processor, implements the steps of a robot control method for adaptive optimization of hand-eye calibration as described above. The self-adaptive optimization robot control method and system for hand-eye calibration have the following beneficial effects: Compared with the hand-eye calibration mode that the sampling path is fixed, a real-time evaluation and feedback mechanism for image quality in the sampling process is lacking and long-term drift is ignored in the prior art, the hand-eye calibration method and the device realize the omnibearing optimization of the calibration process by introducing a bilateral collaborative mechanism combining the self-adaptive optimization path strategy based on online quality feedback and the preset calibration error compensation strategy. The self-adaptive optimization path strategy based on the online quality feedback enables the sampling path to be dynamically adjusted according to the real-time image quality in the data acquisition process, thereby effectively avoiding adverse factors such as illumination interference and shielding, ensuring that the source data participating in calibration calculation has higher accuracy and consistency, and improving the