Dynamics Modeling and Parameters Identification Research on Rope Driven Biped Robot Legs
Zhu Zhenbiao
Tang Jun
Liang Zhiyuan
Li Qingdu
Abstract：The rope driven robot has received widespread attention and application due to the high-effi-ciency conduction and essential flexibility.However,there are problems such as joint coupling,complex dynam-ic models,and difficulty in identifying parameters.To this end,this study proposes an overall identification method based on the rope driven biped robot,using joint angles and current information to overcome the disad-vantages of traditional force sensors,and reduce the difficulty of the overall development of biped robots.Firstly,a complete rope driven dynamic model is established,and the excitation trajectory generated by pattern recogni-tion and algorithm optimization is used.Then,the optimized trajectory is implemented on actual robots,and the joint angles and current information are collected through the sensors.Finally,the least square method with the physical consistency is used to estimate the dynamics parameters.This study uses the M02 rope driven robot to perform dynamic parameter identification experiments and compares the prediction results obtained from the method mentioned with the actual results.The experimental results show that the minimum percentage of the rel-ative error between the predictive torque and the actual torque can reach 13.36％,indicating that the overall identification method based on the joint angle and the current information proposed can be effectively applied to the bilateral robots driven by ropes.
Keywords：Biped robotRope drivenKinematicsDynamicsParameter identification
Publication Date：2024-08-15
Online Publishing Date：2026-09-12(First online date of this platform, not the publication date of the document)
Pages：8( 18-25 )