Trajectory generation and control of a lower limb exoskeleton for gait assistance

• Main Authors: Luo, Lincong, Foo, Ming Jeat, Ramanathan, Manoj, Er, Jie Kai, Chiam, Chye Hsia, Li, Lei, Yau, Wei Yun, Ang, Wei Tech
• Other Authors: School of Mechanical and Aerospace Engineering
• Format: Article
• Language: English
• Published: 2023
• Subjects: Engineering::Mechanical engineering; Dynamic Movement Primitives; Path-based Control
• Online Access: https://hdl.handle.net/10356/170427
• Institution: Nanyang Technological University

The application of lower limb exoskeletons for gait assistance has attracted extensive attention in recent decades. Reference trajectory and assistive control are the important aspects determining the performance of the exoskeleton. During overground walking, there are certain variations in touchdown position, step length and time as subjects have the requirement of adjusting gait pattern to maintain balance. In order to adapt to the different starting and ending foot location on the swing phase of gait, in this study, a dynamic movement primitives (DMPs)-based gait trajectory generation method is proposed to provide a flexible manner to online adjust the reference trajectory. Meanwhile, a time-independent path-based trajectory controller is developed to exert the assistance and guidance for the swing leg motion, where a lookahead distance is employed to search for the desired angular position in 4D hip-knee joint space. The combination of DMP-based reference generation and path-based trajectory control enables lower limb exoskeletons to reduce gait trajectory deviation while still allowing subjects’ own effort to change the gait pattern. Simulations and experiments involving five healthy subjects walking on level ground with a commercial lower limb exoskeleton, were conducted to validate the feasibility and effectiveness of the proposed approach. The results showed that the exoskeleton assistance improved subjects’ leg movement performance, and did not limit subjects’ active effort to adjust step length and time.