Adaptive Torque Control of Exoskeletons under Spasticity Conditions via Reinforcement Learning

2025-03-14

AI Summary

For patients with spastic movement disorders—such as those with cerebral palsy or post-stroke sequelae—knee exoskeletons pose significant safety risks and challenges in individualized control. This paper proposes a deep reinforcement learning framework integrating a differentiable spastic reflex model with a muscle–exoskeleton digital twin. It is the first approach to embed Modified Ashworth Scale–informed spasticity characteristics into a differentiable biomechanical model, while jointly optimizing task performance and human–robot interaction forces. Compared with conventional compliant controllers, the proposed method reduces peak joint torque by 10.6% on average in both simulation and physical experiments, and decreases the root-mean-square interaction force during the settling phase by 8.9%. These improvements markedly enhance control safety and adaptability for heterogeneous spastic populations.

Technology Category

• Intelligent Robots: Behavior Learning & Control
• Natural Language Processing: Safety and Robustness
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Spasticity is a common movement disorder symptom in individuals with cerebral palsy, hereditary spastic paraplegia, spinal cord injury and stroke, being one of the most disabling features in the progression of these diseases. Despite the potential benefit of using wearable robots to treat spasticity, their use is not currently recommended to subjects with a level of spasticity above ${1^+}$ on the Modified Ashworth Scale. The varying dynamics of this velocity-dependent tonic stretch reflex make it difficult to deploy safe personalized controllers. Here, we describe a novel adaptive torque controller via deep reinforcement learning (RL) for a knee exoskeleton under joint spasticity conditions, which accounts for task performance and interaction forces reduction. To train the RL agent, we developed a digital twin, including a musculoskeletal-exoskeleton system with joint misalignment and a differentiable spastic reflexes model for the muscles activation. Results for a simulated knee extension movement showed that the agent learns to control the exoskeleton for individuals with different levels of spasticity. The proposed controller was able to reduce maximum torques applied to the human joint under spastic conditions by an average of 10.6% and decreases the root mean square until the settling time by 8.9% compared to a conventional compliant controller.

Problem

• Develop adaptive torque control for exoskeletons under spasticity.
• Address varying dynamics of spasticity in wearable robots.
• Reduce interaction forces and improve task performance in spastic conditions.

Innovation

• Deep reinforcement learning for adaptive torque control
• Digital twin with musculoskeletal-exoskeleton system
• Differentiable spastic reflexes model for muscle activation