Model-based predictive control of wheeled omni-directional robots considering nonlinear dynamical constraints and input delay

Abstract

The main enthusiasm of the use of MPC in this study relies on its ability in including state and control constraints that naturally comes up in practical problems. At the same time, nonlinearity of constraints compels the numerical solution of optimization problem. Therefore, a number of intelligent algorithms to overcome nonlinear optimization of the cost function is reviewed, however, their time consuming computations are considered as a dominant limitation. Meantime, due to its remarkable advantages, the motion control of Omni-directional mobile robot is handled by design of NMPC scheme. Furthermore, a novel kinematic model for the application of mobile robot is represented. As a demonstration for the proposed model, simulation results show that settling time is reduced 22% in comparison with ideal kinematic model and also the performance of the suggested control strategy shows its effectiveness in presence of disturbance and noise in all environments.