A State Graph Generation and Control Optimization Method for Vision-Based Collaborative Robot

In recent years, research on collaborative robots based on visual technology has experienced rapid development. Machine vision-based robot control and applications have become important research topics in this interdisciplinary field. Control optimization of robots is a key focus in this domain, as many robots still exhibit redundant operations during their functioning. This paper takes a vision-based collaborative Rubik's Cube Solving Robot as an example and proposes an algorithm for generating a state graph based on machine vision recognition. This algorithm forms the basis for an optimal search for the Rubik's Cube solving path. The paper designs and fabricates a two-gripper collaborative robot using machine vision establishes corresponding kinematic and physics simulation models, and conducts experiments to test the proposed algorithm. Simulation and prototype experiments demonstrate that the proposed algorithm can optimize the steps for the robot to solve the Rubik's Cube, thereby enhancing the overall work efficiency of the collaborative robots.