A Collaborative Action Sequence Planning Method for Dual-arm Piano Playing Robot Arm-Hand Based on Elite retention Multi-objective Optimization

As a typical representative of humanoid robots showing the dexterous operation ability of dual-arm, the research on the piano playing robot can promote the rapid development of humanoid robot applications. In order to solve the problem of insufficient rhythm and integrity of the performance caused by the difficulty of arm-hand coordination in the process of playing the piano, a collaborative action sequence planning method for dual-arm Piano playing robot based on elite retention Multi-objective optimization is proposed. Firstly, a piano playing robot arm-hand motion model based on performance effects is established by analyzing the arm-hand motion states during human piano playing and combining professional performance evaluation metrics. Secondly, the elite retention strategy and the Rayleigh distribution multi-objective particle swarm optimization algorithm are used to improve the optimal piano action sequence for arm-hand cooperation. Finally, taking the piano piece "Ode to Joy" as an example, the algorithm proposed in this paper is used to test the performance of the two-arm robot, and the experimental results show that the method has a good global optimization ability in solving the action sequence planning problem of the arm-hand, and can improve the rhythm of the robot's repertoire.