Multivariable grasping force control of myoelectric multi-fingered hand prosthesis

Dutra, Bruno Gomes,Silveira, Antonio da S.

DOI: https://doi.org/10.1007/s40435-023-01130-8

2023-03-04

International Journal of Dynamics and Control

Abstract:This article presents a novel force control strategy to manipulate and grasp objects with multi-fingered hand prostheses. For this purpose, we develop an approach based on the multivariable state-space technique, which uses the Kalman filter and the model predictive control for real-time optimal grasping force control and slippage prevention. We utilized a low-cost prosthetic hand and the least-squares algorithm to identify the multiple-input and single-output state-space force model. The Kalman filter design uses this force model and the fusion of the prostheses' force sensors to improve the global grasping force estimation by the redundancy effect. Then, we use the model predictive control to exploit the optimal grasping force control and perform anthropomorphic movements with the prosthesis. The results have demonstrated the feasibility of the approach to track force references in real time and provide force control with everyday objects. The performance of the proposed approach was superior to the results obtained with the decentralized PID control used in the state-of-the-art. The results confirmed that the proposed grasping force method is simple, efficient, and can be easily embedded in micro-controlled systems to be applied into commercially available prostheses as well as into low-cost prostheses.