Session: 07-10-02: Medical Robotics, Rehabilitation, and Surgery II

Paper Number: 165210

Motion Planning and Control of Lower Limb Exoskeletons to Perform Rehabilitation Exercises After Total Knee Arthroplasty 

Lower limb exoskeletons (LLEs) have emerged as a promising rehabilitation tool for assisting individuals recovering from musculoskeletal impairments, particularly patients undergoing Total Knee Arthroplasty (TKA). These robotic systems provide controlled assistance and resistance during movement, promoting neuromuscular recovery and improving mobility. However, challenges remain in developing control strategies that ensure both adaptive gait assistance and postural stability, which are critical for safe and effective rehabilitation. This research focuses on the development and implementation of a control system for the EXO-H3 exoskeleton using MATLAB Simulink, aimed at enhancing rehabilitation exercises for TKA patients.

The proposed control framework is grounded in Central Pattern Generator (CPG) dynamic equations, which are widely used for generating biologically inspired rhythmic movements. The CPG-based approach enables the exoskeleton to produce smooth, natural gait trajectories by adjusting key parameters such as gait frequency, oscillation amplitude, equilibrium position, and phase variation. To further refine trajectory generation, Fourier series representations are employed, allowing for precise modulation of movement patterns and improved adaptability to patient-specific needs. Additionally, real-time feedback from interaction torque and motor torque is integrated to enhance control accuracy, ensure patient safety, and facilitate a responsive rehabilitation process.

Beyond trajectory generation, this research seeks to improve the postural stability of LLEs by dynamically modifying hip and ankle trajectories in real-time. Maintaining balance is crucial for exoskeleton users, particularly during rehabilitation, where even minor deviations from stable posture can lead to falls or inefficient gait patterns. To address this challenge, this study employs Divergent Component of Motion (DCM)-based control strategies. The DCM framework provides a mathematical representation of balance dynamics, enabling predictive adjustments to lower limb trajectories that prevent instability. Both time-independent DCM-based control (TIC) and time-dependent DCM-based control (TDC) strategies are implemented to minimize DCM error and optimize postural stability during gait cycles. The TIC strategy ensures long-term stability by maintaining a bounded DCM trajectory, while the TDC strategy dynamically adjusts gait parameters in response to real-time variations in user movement and external disturbances.

The methodology involves extensive simulation and validation of the proposed control system in MATLAB Simulink. Various gait conditions and rehabilitation scenarios are simulated to assess the effectiveness of the CPG-based trajectory generation and DCM-based stability control. Performance metrics such as trajectory tracking accuracy, torque responses, DCM error minimization, and overall gait stability are analyzed to evaluate the system’s effectiveness. By incorporating adaptive control mechanisms that account for both motion generation and postural stability, this research aims to create a robust exoskeleton control system that enhances rehabilitation outcomes for TKA patients.

This study contributes to the broader field of assistive robotics by advancing control strategies that improve both mobility and stability in lower limb exoskeletons. The integration of biologically inspired trajectory generation with real-time postural control presents a novel approach to optimizing rehabilitation assistance. Future work will involve experimental validation on human subjects to further refine the control strategies and assess their efficacy in real-world rehabilitation settings. Ultimately, this research aims to bridge the gap between robotic control theory and clinical rehabilitation, paving the way for more effective, adaptable, and user-friendly exoskeleton technologies.

Presenting Author: Mojtaba Sharifi San Jose State University

Presenting Author Biography: Mojtaba Sharifi has been an Assistant Professor at the Department of Mechanical Engineering, San Jose State University, San Jose, California, USA, since January 2022. He leads research projects on the design, control, and autonomy of assistive and rehabilitation robotics including lower-limb and upper-limb exoskeletons for safe, compliant, and intelligent interaction with humans.
Prior to this, he was a Postdoctoral Research Fellow working on autonomous control of physical human-robot interaction (pHRI), medical robotics, haptics, collaborative robotics, tele-robotics, and assistive technology at the University of Alberta, Canada. He was with the Department of Electrical and Computer Engineering and the Department of Medicine.

Authors:

Pinqian Lin San Jose State University
Mojtaba Sharifi San Jose State University