Vector polynomial based analytical inverse kinematics and configuration design of 6R robotic arms

• Published in: Communications in nonlinear science & numerical simulation 2025-03, Vol.142, p.108546, Article 108546
• Main Authors: Chen, Feifei, Rui, Xiaoting, Ju, Hehua, Wang, Guoping, Wang, Kaimeng
• Format: Article
• Language: English
• Subjects: 6R robotic arm; Configuration design; Inverse kinematics; Vector polynomial
• ISSN: 1007-5704
• DOI: 10.1016/j.cnsns.2024.108546

•Formulate the kinematic model of 6R robotic arms into a vector polynomial system.•Design robot configurations to decouple position and attitude in the kinematic model.•Improve the Dixon resultant method for symbolic processing and numerical computation.•Develop software for inverse kinematic modelling and computation of robotic arms. Robotic arms with high precision play a crucial role in industries like aerospace and intelligent manufacturing. Currently, industrial robotic arms typically adopt spherical wrists to obtain analytical solutions; however, manufacturing-related structural errors significantly reduce the end effector's absolute positioning accuracy, impeding the development towards high-precision autonomous control. This study presents a series of kinematic modelling, configuration design, and analytical inverse solutions for robotic arms with the goal of delivering precise and effective inverse kinematic solutions, while taking the engineering limitation into consideration. Taking the vector polynomial system as the key premise, an iterative kinematic model with the fewest number of equations, independent variables, and orders of magnitude is then established. Through decoupling analysis of the position and attitude in the kinematic model, configuration designs of robotic arms with orthogonal or parallel axes are proposed without adhering to the restrictions of three-axis intersection. Additionally, the inverse kinematic issue is addressed by conversion into polynomial space using the Dixon elimination approach. Eight sets of inverse kinematic solutions are obtained for any reachable position and attitude in simulation. The computation time does not exceed 2 milliseconds, and both position and attitude relative errors are below 10–15, enhancing computational accuracy for robot kinematics. The work of this paper provides a practical kinematic theory for the development of robotic arms with high absolute positioning accuracy and efficiency.