Trajectory tracking of mobile robots based on model predictive control using primal dual neural network

Abstract

Mobile robots' motion is constrained by the maximum velocity its actuators can provide, when it tracks a reference trajectory which imposes demanding requirements on the robot's driving capabilities. In this paper, a model predictive control (MPC) scheme is proposed for trajectory tracking control of two-wheel mobile robots. Based on the derived tracking-error kinematics of the robot, the proposed MPC approach can be iteratively formulated as a quadratic programming (QP) problem, which can be solved using a linear variable inequality based primal-dual neural network (LVI-PDNN) over a finite receding horizon. The applied neural networks are both stable in the sense of Lyapunov and globally convergent to the exact optimal solutions of reformulated convex programming problems. The smoothness of the robot motion is improved, a reasonable magnitudes of the robot velocities and a better tracking performance are achieved. Simulation and experimental results are provided to demonstrate the effectiveness and characteristics of the proposed LVI-PDNN based MPC approaches to trajectory tracking control.