Model predictive control for visual servo steering of nonholonomic mobile robots

In this paper, model predictive controller for the visual servo steering of a mobile robot in polar coordinate is presented. Firstly, a kinematic predictive steering controller utilized to generate the command of velocity is introduced. Then, a dynamic predictive controller is designed to steer the system. The model predictive control (MPC) can deal with the constraints easily and it can be iteratively formulated as a quadratic programming (QP) problem, which can be solved using neurodynamic optimization, for example, primal-dual neural network (PDNN), over a finite receding horizon. The applied neurodynamics are globally convergent to the exact optimal solutions of reformulated convex programming problems. Finally, experiment results are provided to illustrate the effectiveness of the MPC scheme.