Robot Probe Self-Calibration for Accurate Kinematic Parameters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robot calibration methods rely on external measurement devices, leading to higher costs, longer time consumption, and lower accuracy due to environmental factors affecting the accuracy of kinematic parameter calibration.
Innovation Solution
A robot calibration method that establishes kinematic models between calibration blocks and the robot's probe end, determining pose data and contact coordinates to measure the robot's workspace automatically, without external apparatuses, using sensors to ensure accurate contact strength and correcting kinematic parameters through closed-loop calibration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If external measurement devices are used for robot calibration, then measurement can be performed, but accuracy is reduced due to environmental factors and device complexity increases
Solution Approach 1:
The robot uses its own probe end to measure calibration blocks and determine its own kinematic parameters, eliminating the need for external measurement devices. The robot system performs self-calibration by measuring contact coordinates between the probe end and calibration blocks, then correcting its own geometric parameters based on measurement deviations.
Solution Approach 2:
The patent extracts and removes the external measurement device from the calibration system, replacing it with the robot's own measurement capabilities. By taking out the problematic external device that introduces environmental errors, the system achieves higher accuracy through self-measurement using the probe end.
2Measurement precision
If external measurement devices are used for calibration, then measurement can be performed, but time consumption and cost increase
Solution Approach 1:
The robot performs self-calibration using its own probe end and onboard sensors, eliminating the need for external measurement devices and the associated setup time. The automated self-measurement process reduces calibration time while improving accuracy by removing environmental interference from external devices.
Solution Approach 2:
The patent replaces the mechanical external measurement device system with an automated sensor-based measurement system integrated on the robot. This substitution eliminates the need for complex external apparatus setup and reduces calibration time through automated measurement and calculation processes.
3Measurement precision
If traditional calibration methods are used, then calibration can be performed, but the entire workspace cannot be measured and accuracy is limited
Solution Approach 1:
The patent employs dynamic calibration where the robot moves the probe end to multiple positions and orientations throughout its entire workspace, measuring calibration blocks at various locations. This dynamic approach allows comprehensive coverage of the workspace volume, unlike static traditional methods that are limited to specific measurement points.
Solution Approach 2:
The patent extends calibration from traditional limited-point measurement to three-dimensional workspace coverage by utilizing the probe end's ability to reach and measure calibration blocks throughout the entire operational volume. This adds spatial dimensionality to the calibration process, ensuring accuracy across all workspace regions.
Data Source
AI summary
Provided are a robot calibration method and apparatus, an electronic device and a storage medium. The method comprises: establishing an i-th kinematic model between a first coordinate system of an i-th calibration block and a second coordinate system of a detection end of a robot, wherein i is a positive integer greater than or equal to 1 and less than or equal to n, and n is the number of calibration blocks; determining a plurality of pieces of pose data of the detection end in contact with a plurality of test points on the i-th calibration block; determining a plurality of contact coordinates of the detection end according to the plurality of pieces of pose data and the i-th kinematic model; and determining a measurement parameter of the robot according to n groups of contact coordinates corresponding to the n calibration blocks.