Development and Repetitive Learning Control of Lower Limb Exoskeleton Driven by Electrohydraulic Actuators

• Published in: IEEE transactions on industrial electronics (1982) Vol. 64; no. 5; pp. 4169 - 4178
• Main Authors: Yang, Yong, Ma, Lei, Huang, Deqing
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.05.2017; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Actuators; Backstepping; CASWELL; embedded electronic system; Embedded systems; Exoskeletons; Learning; Legged locomotion; Limbs; lower limb exoskeleton; microcontroller; Microcontrollers; repetitive learning control (RLC); Rigid-body dynamics; Sensors; Tracking; Tracking control; Valves
• ISSN: 0278-0046; 1557-9948
• DOI: 10.1109/TIE.2016.2622665

This paper presents a novel development of lower limb exoskeleton, which is named as CASWELL, and its repetitive learning control design for movement assistance. CASWELL has five degrees of freedom per leg, two of which are driven by linear single-rod electrohydraulic actuators. First, the mechanical structure and embedded electronic system are constructed, respectively. Second, by incorporating the systems of rigid body and electrohydraulic actuators, the complete dynamics of CASWELL are analyzed and modeled. Third, under the framework of backstepping design, a repetitive learning control scheme is presented to address the periodic tracking control of lower limbs of exoskeleton, where the learning convergence of the closed-loop system is proved rigorously in a Lyapunov way. Finally, the proposed controller is implemented in the embedded electronic system via a 32-bit microcontroller, and tested on the developed CASWELL. The experimental results are given to demonstrate the performance of the whole exoskeleton system.