Related Experiment Video
Updated: Nov 19, 2025
11:16
Engineering Platform and Experimental Protocol for Design and Evaluation of a Neurally-controlled Powered Transfemoral Prosthesis
Published on: July 22, 2014
16.5K
Evaluation of Motor Primitive-Based Adaptive Control for Lower Limb Exoskeletons
Polyana F Nunes1, Icaro Ostan1, Adriano A G Siqueira1
1Rehabilitation Robotics Laboratory, Department of Mechanical Engineering, São Carlos School of Engineering, University of São Paulo, São Carlos, Brazil.
Frontiers in Robotics and AI
|January 27, 2021
Summary
This study introduces a bio-inspired controller for lower limb exoskeletons, enhancing stroke rehabilitation by adapting to user needs and compensating for reduced torque during walking. The controller effectively assists movement and adapts to gait changes.
Area of Science:
- Rehabilitation Robotics
- Biomechanical Engineering
- Neurorehabilitation
Background:
- Lower limb exoskeletons and advanced control strategies are crucial for post-stroke movement rehabilitation.
- Existing robot-assisted therapy offers support and adaptability but has shown limited effectiveness.
- Bio-inspired control approaches are being explored to improve exoskeleton performance.
Purpose of the Study:
- To develop and evaluate a novel bio-inspired controller for lower limb exoskeletons based on motor primitives.
- To assess the controller's ability to assist post-stroke individuals in movement rehabilitation.
- To investigate the controller's effectiveness in compensating for user torque deficiencies and adapting to gait variations.
Main Methods:
- A bio-inspired controller utilizing motor primitives was developed.
- User torques were estimated using a generalized momentum-based disturbance observer and an extended Kalman filter.
- The controller was evaluated on a lower limb exoskeleton with a series elastic actuator at the knee joint during active, passive, and combined walking conditions.
Main Results:
- The robot assistance successfully compensated for motor primitive weight deficiency when subjects exerted less torque than expected.
- Robot torque applied at the knee influenced motor primitive weights related to the hip joint, despite only the knee being actuated.
- The system demonstrated adaptation to changes in gait characteristics within three to four steps and compensated for asynchronous movements.
Conclusions:
- The proposed bio-inspired controller shows promise for enhancing robot-assisted gait rehabilitation in post-stroke individuals.
- The controller's ability to adapt and compensate for user deficits suggests improved therapeutic outcomes.
- This approach offers a more personalized and effective rehabilitation experience by leveraging motor primitive concepts.
