Robust tracking control of a minimal realization model of an impaired human hand for anthropomorphic coordination

Abstract

Objective Physiologically relevant optimal controllers better represent the decision-making process of the central nervous system (CNS) with proper neural inputs and proprioceptor feedback. A biomechanical mathematical framework in the human palm reference frame was simulated using physiological dynamics to explore the biomechanics of movement coordination of fingers in a human hand.Method Physiological state space models include multiple zero eigenvalues, representing a redundant system. A fingertip trajectory tracking control paradigm is created by reducing the model order through the H∞ control paradigm. The external disturbances and sensory noise are included in the 21-Degrees-of-Freedom biomechanical model.Main Results An analysis is conducted on the flexion movement of the robotic finger when it is disrupted from its initial equilibrium position. The controller administers feedback force at the joints to control the movement of the robotic finger. A full order H∞ robust controller was developed, and the results were compared with the reduced-order model. The model order was reduced from 21 to 18 states. A reference trajectory is followed by the index finger in the minimal realization model, which is derived from the joint angular position profile. It stabilizes at a flexion angle of 1 rad/sec within 2 sec. Studying the simpler form of a model first gives a realistic view of the more complex model. This study has the potential to enhance our comprehension of anthropomorphic movement coordination in hands with impairments in kinesiology, ergonomics, assistive technology, and prosthetic devices.

Keywords

• Hand biomechanical model
• Central nervous system
• Minimal realization
• Tracking control