Title: Model adaptive collision detection for flexible joint manipulator based on state observer

Authors: Cui Shipeng; Liu Yiwei; Sun Yongjun; Chris Gerada

Addresses: State Key Laboratory of Robotics and System, Harbin Institute of Technology, Harbin 150001, China ' State Key Laboratory of Robotics and System, Harbin Institute of Technology, Harbin 150001, China ' State Key Laboratory of Robotics and System, Harbin Institute of Technology, Harbin 150001, China ' Department of Electrical Engineering, University of Nottingham, UK

Abstract: In this paper, a novel collision detection method for flexible joint manipulator, which calculates the difference between theoretical control torque of regression model and measured control torque, is proposed. As the core of the collision detection method, the adaptive law based on regression model error improves the accuracy of collision detection and makes the collision detection have a first-order low-pass characteristic. Specifically, it assumes that the states of flexible joint manipulator can be accurately estimated by employing the state observer based on fixed-lag Kalman smoother. Both simulations and experiments are carried out on a 2-degree of freedom (DoF) flexible joint manipulator, and the results demonstrate that the proposed method can notably improves collision detection accuracy and sensitivity.

Keywords: collision detection; flexible joint manipulator; adaptive law; Kalman smoother.

DOI: 10.1504/IJHM.2023.132298

International Journal of Hydromechatronics, 2023 Vol.6 No.3, pp.242 - 257

Received: 09 Feb 2022
Accepted: 07 Apr 2022
Published online: 17 Jul 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article