Research on Dynamic Model of Flexible-Jointed Space Manipulator Based on Gaussian Process

• Published in: Chinese Automation Congress (Online) pp. 2239 - 2244
• Main Authors: Liu, Changhao, Gong, Tengshang, Pan, Fei, Gong, Sijin, Hua, Gengpai
• Format: Conference Proceeding
• Language: English
• Published: IEEE 25.11.2022
• Subjects: Analytical models; dynamic model; Dynamics; flexible-jointed space manipulator; Gaussian process; Gaussian processes; Heuristic algorithms; Predictive models; semi-parametric model; Training; Uncertainty
• ISSN: 2688-0938
• DOI: 10.1109/CAC57257.2022.10055959

Establishing an accurate dynamic model is the key to realize the high-precision motion and control of manipilator. The space manipilator is always affected by various uncertainty problems, so the dynamic model often deviates as a result. Therefore, this paper takes the floating-based flexible jointed space manipilator as the object, and carries out the research of manipilator dynamic modeling algorithm for the uncertainty problem of dynamic modeling, focusing on the analysis of the parameter-based space manipilator dynamic modeling method and the Gaussian process-based space manipilator dynamic modeling method. On this basis, proposes a semi-parametric dynamic modeling method combining the two models, and the three algorithms Simulation experiments and comparative analysis were conducted to verify the feasibility of the Gaussian process-based dynamic modeling method for solving the dynamics uncertainty problem.