Journal of Bionic Engineering ›› 2024, Vol. 21 ›› Issue (4): 1761-1774.doi: 10.1007/s42235-024-00534-2
A Learning‑based Control Framework for Fast and Accurate Manipulation of a Flexible Object
Junyi Wang1,2; Xiaofeng Xiong2 ; Silvia Tolu1 ; Stanislav N. Gorb3
- 1 Department of Electrical and Photonics Engineering, Technical University of Denmark, Anker Engelunds Vej 101, Kongens Lyngby, Zealand 2800, Denmark 2 SDU Biorobotics, M?rsk Mc-Kinney M?ller Institute, University of Southern Denmark (SDU), Campusvej 55, Odense M, Funen 5230, Denmark 3 Department Functional Morphology and Biomechanics, Zoological Institute, University of Kiel, Am Botanischen Garten 1–9, D-24118 Kiel, Schleswig-Holstein, Germany
A Learning‑based Control Framework for Fast and Accurate Manipulation of a Flexible Object
Junyi Wang1,2; Xiaofeng Xiong2 ; Silvia Tolu1 ; Stanislav N. Gorb3
- 1 Department of Electrical and Photonics Engineering, Technical University of Denmark, Anker Engelunds Vej 101, Kongens Lyngby, Zealand 2800, Denmark 2 SDU Biorobotics, Mærsk Mc-Kinney Møller Institute, University of Southern Denmark (SDU), Campusvej 55, Odense M, Funen 5230, Denmark 3 Department Functional Morphology and Biomechanics, Zoological Institute, University of Kiel, Am Botanischen Garten 1–9, D-24118 Kiel, Schleswig-Holstein, Germany
摘要： This paper presents a learning-based control framework for fast (< 1.5 s) and accurate manipulation of a fexible object, i.e., whip targeting. The framework consists of a motion planner learned or optimized by an algorithm, Online Impedance Adaptation Control (OIAC), a sim2real mechanism, and a visual feedback component. The experimental results show that a soft actor-critic algorithm outperforms three Deep Reinforcement Learning (DRL), a nonlinear optimization, and a genetic algorithm in learning generalization of motion planning. It can greatly reduce average learning trials (to < 20% of others) and maximize average rewards (to > 3 times of others). Besides, motion tracking errors are greatly reduced to 13.29% and 22.36% of constant impedance control by the OIAC of the proposed framework. In addition, the trajectory similarity between simulated and physical whips is 89.09%. The presented framework provides a new method integrating data-driven and physics-based algorithms for controlling fast and accurate arm manipulation of a fexible object.