ICMO-AFS: Robotic Comprehensive Motion Optimization Based on Incremental Accelerated Gradient Descent and Adaptive Fine-Tuning Strategy
Shengjie Li,Jin Wang,Haiyun Zhang,Yichang Feng,Guodong Lu,Anbang Zhai
DOI: https://doi.org/10.2139/ssrn.4379717
2023-01-01
Abstract:In the robot intelligent manufacturing scenario, working efficiency, energy consumption and motion stability are the important factors to evaluate the robotic working performance. In this study, the problem of collision-free comprehensive motion optimization for serial robots is discussed. Aiming at the Time, Energy consumption, and Impact (T-E-I) of the manipulator during movement, an incremental comprehensive motion optimization algorithm with adaptive fine-tuning strategy (iCMO-AFS) is proposed. Firstly, the objective function is constructed for the comprehensive performance optimization problem, and the constraints are converted into penalty functions to facilitate the solution. In particular, for the modeling problem of robot energy consumption, the static power regression model is constructed to improve the evaluation accuracy. Secondly, based on the optimization framework of adaptive sampling and accelerated gradient descent, the incremental trajectory optimization of the T-E-I performance is carried out. In addition, the above method is extended to the velocity optimization problem, which further improves the comprehensive motion performance of the robot. Thirdly, an adaptive fine-tuning strategy is proposed to modify the sampling method and collision model, which improves the solution performance. Simulation and experimental tests in several scenarios show that the proposed method can achieve satisfactory results in both planning efficiency and working performance.