Sensor-Fusion-Driven Adaptive Control of Elastic Actuation-Based Wearable Exoskeleton

This paper proposes an adaptive control framework for a compliant lower limb exoskeleton that assists the motion of the hip and knee joints of the sagittal plane. The system leverages real-time multisensor fusion, virtual joint elasticity, and an adaptive fuzzy PID control strategy to achieve robust trajectory tracking under dynamic and user-specific variability. The framework is validated through high-fidelity simulations and hardware-in-the-loop experiments, demonstrating improved tracking precision, motion smoothness, and control responsiveness compared to conventional PID schemes. The integration of virtual compliance and intelligent gain adaptation enhances interaction safety and comfort, supporting the practical deployment of adaptive control in wearable robotic systems for rehabilitation and mobility assistance.