Robust and compliance control for robotic knee prosthesis using admittance model and sliding-mode controller

Abstract

Achieving compliance and flexibility under the premise of ensuring trajectory tracking performance and also reflecting the wearer’s movement intention, has not yet been well solved in the field of prosthesis. The aim of this paper is to provide a compliant, robust, and continuous control scheme for robotic knee prosthesis to solve the contradictory problems of trajectory tracking performance and compliance. The proposed scheme are based on the admittance model and radial basis function (RBF) neural network–enhanced nonsingular fast terminal sliding-mode controller (NFTSMC). The desired trajectory of the prosthetic knee joint is driven by humans and reshaped to reference trajectory by an admittance model, so that the prosthetic leg can reflect the human’s movement intention and being compliant. RBF neural network is introduced to achieve adaptive approximation of unknown models and ensure that the controller does not depend on the mathematical model of the “human-in-the-loop” prosthesis system. A novel NFTSMC was proposed to deal with the influence of ground reaction forces (GRFs) and fitting errors of the RBF neural network, which make the tracking error converge to zero in a finite time. The adaptive law of the RBF neural network is obtained by the Lyapunov method, and the stability and finite-time convergence of the closed-loop system are rigorously proved and analyzed mathematically. The simulation results prove the feasibility and effectiveness of the propose control scheme.