Parameterized Model Predictive Control of a Nonholonomic Mobile Manipulator: A Terminal Constraint-Free Approach

Abstract

In this paper, a non-linear model predictive control of a mobile manipulator, consisting of a differential-drive mobile base and a 6 DOF arm, is presented. A novel approach to solve the online optimization problem that reduces the overall computational time is proposed, focusing on solving the trajectory tracking problem for handling and grasping operations of a mobile manipulator. The novelty is the introduction of parameterized control inputs and removing terminal constraints. Experimental results comparing our method with traditional non-linear MPC schemes are also reported. They show that the controller is able to achieve an average Cartesian error on the end-effector position of 2 cm.