A Vision-Based Simultaneous Calibration Method for Dual-Robot Collaborative System

Abstract

Dual-robot collaborative system has shown a significant potential for performing complex tasks. However, excellent interaction between cooperative manipulator involves a wide variety of coordinate frame transformations. Therefore, it is crucial for a dual-robot collaborative system to solve transformation relationships among the coordinate frames. In this article, a generalized matrix equation <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">AXP_B</i> = <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">YCP_Z</i> is formulated to described the kinematic transformation relationships in a dual-robot collaborative system. Furthermore, a novel vision-based simultaneous calibration method, consisting of a closed-form solution for estimation and an iterative optimization based on Lie algebra, was proposed for solving these unknown transformation matrices. It is worth that the proposed kinematic equation and calibration method only relies on the position information of the marker and provides greater versatility. Finally, in order to confirm the advancement of the proposed method, three mainstream methods are chosen for comparison. Relevant simulations and experiment demonstrate the novel calibration method provides great superiority in the accuracy, robustness, efficiency, and scenario compatibility, which can be applied to a variety of dual-robot collaboration tasks.