Lane keeping control on Mecanum wheeled omnidirectional vehicles using laser scanner

Abstract

In recent years, there is a strong demand for automated guided vehicles in many fields such as manufacturing or studying and a variety of omnidirectional vehicle systems has been developed for these purposes, which includes the Mecanum-wheel vehicles. In the meantime, identifying the environment by laser scanner has increasingly become an efficient solution for robotic recognition. In order to implement an intelligent vehicle for industrial environment, a lane keeping method for Mecanum wheeled Omni-directional vehicles is proposed in this paper. We first design a LQR controller for the omnidirectional vehicle motion control. After that, a laser scanner is used to sense the lane for the vehicle. In the aspect of algorithm, a set of waypoints is used to determine the heading direction of the vehicle when cornering and a PID controller is designed to keep the vehicle in the detected lane when it is passing a relatively straight lane. The simulation experiment shows that the omnidirectional platform tracking system has a good response by using PID control. Finally, experimental result shows that the real vehicle can track the desired trajectory rapidly and smoothly, and the vehicle meanwhile is constrained in the lane in the whole process.