Role Dynamic Assignment of Human–Robot Collaboration Based on Target Prediction and Fuzzy Inference
Chengyun Wang, Jing Zhao
Abstract
Aiming to solve the problem of role dynamic assignment in human–robot collaborative motion with unknown and changeable targets, a role dynamic assignment method based on target prediction and fuzzy inference is proposed. First, a human–robot collaborative motion framework based on role dynamic assignment is constructed. Then, based on the analysis of the human–robot collaborative motion with fixed role (HMFR), a human–robot collaborative motion target prediction module, a role dynamic assignment module, and a robot motion planning module are designed to realize the role dynamic assignment and motion planning of the robot. Finally, human–robot collaborative motion experiments based on role dynamic assignment are performed. Experimental results show that the proposed role dynamic assignment algorithm can effectively implement role dynamic assignments. Compared with the human–robot collaborative motion with fixed roles, the human–robot collaborative motion based on role dynamic assignment has better overall performance in terms of actual path and desired path similarity (Fréchet distance) and participant workload.