Model predictive control for a musculoskeletal robot with motor driven artificial muscle

Abstract

This work develops the model predictive control (MPC) for an upper limb musculoskeletal robot model. Firstly, the numerical relationship between muscle force and bone joints is obtained, and then the control structure is designed on the basis of the joint torque and joint angles. Due to the nonlinearity of the control object, we linearize it using a feedback linearization method. Secondly, a model predictive controller is designed to obtain the linearized model. By analyzing the second-order closed-loop control system, the parameters can be adjusted to obtain a better performance. The control goal of this paper is to make a musculoskeletal robot track the desired trajectory. Simulation results verify that the designed controller is validated.