Motion Control of Wheeled Mobile Robots with Model Predictive Techniques
L. Pacheco, Na Luo · 2014
This work outlines MPC (model predictive control) studies for mobile robots when local motion is considered. Different kinematic mobile robot configurations are introduced. Experiences are conducted by using a differential driven WMR (wheeled mobile robot) as an available vehicle that can represent nonholonomic kinematics constraints. Motion control is done using dynamic models, and selecting orientation and local desired-point distances as control law parameters. MPC is implemented by computing the horizon of suitable coordinates. Command speeds corresponding to the desired point are obtained by minimizing a cost function in which the population of available coordinates is taken into account. The research reported in this paper includes a set of experimental results obtained by testing different trajectories. Parameters such as time, accuracy, speed, and travelled distance are compared from statistics for different control laws.