Intelligent Control of Exoskeletons for Human Limbs using Knowledge-Based Fuzzy Inference System

This paper presents the development of a fuzzy inference system for generating control signals to drive an upper limb exoskeleton using electromyography (EMG) data. The main objective is to create a model that effectively mimics the behavior of the human biological system. Fuzzy logic is employed to capture the relationship between EMG data and elbow angle, enabling the creation of a comprehensive model that accurately represents the intricate movements of the human upper limb. A knowledge base is established to incorporate user preferences and comfort levels, and the fuzzy inference system is designed to replicate the decision-making capabilities of the exoskeleton based on the knowledge base inputs. The generated control signals, derived through fuzzy logic, drive the exoskeleton's actuator. Experimental results demonstrate the system's ability to accurately generate trajectories for actuator movement. To validate the system's accuracy, the actuator's dynamics characteristics are compared with the forearm flexion characteristics once the actuator is driven using the control signals.