Development Of An Advanced Control Laws For A Mobile Manipulator Robot

Bouzoualegh,  Samir; Guechi, El-Hadi

Development Of An Advanced Control Laws For A Mobile Manipulator Robot

Date

2021

Authors

Bouzoualegh, Samir

Guechi, El-Hadi

Publisher

University 20 August 1955 – Skikda

Abstract

In recent years, the control of mobile manipulator robots has been the subject of much research, due to the robots’ increasingly frequent use in dangerous or inaccessible environments, where human beings can hardly intervene. These robot models are highly nonlinear which makes the control strategy very difficult. This obliges us to offer new advanced control for controlling this type of robots, hence in this thesis we propose a new control approach based on a model predictive control (MPC) for controlling a manipulator mobile robot. The new proposed approach of control is a combination of a feedback linearization control (in order to transform the nonlinear dynamics of the system to one completely or partially linear) and an MPC control (in order to obtain better performance for the system), which constitutes the main contribution of this thesis. The control strategy of this robot is done in the following way: first we have to control the mobile base, so that it arrives ata position where the objective fits into the working space of a manipulator arm. In the second step, we must control the manipulator arm so that its end effector achieves its objective. In our work, to overcome the nonlinearity problem, we have used the feedback linearization control for the manipulator arm and Input-output linearization control for the mobile base. Once, the linear models are obtained, a model predictive control approach is developed. We have introduced a quadratic criterion and these parameters are calculated to have a specific behavior of the closed loop system. The control approach design was based on the dynamic models of the robots and we used it to control several manipulator robots and we also used it to control a differential-drive mobile robot (DDMR). The obtained simulation results show the efficiency of the proposed approach of control.