Research On AGV Localization And Path Following Technology Based On Mecanum Wheels

An automated guided vehicle(AGV)is an unmanned transport vehicle that follows a given path and have certain safety protection methods.In recent years,the Mecanum wheeled AGVs have been increasingly applied to various fields such as industrial production,warehousing and exploration of other planets because of their flexibility and efficiency.Researches on AGV have been focused on the following three aspects: multi-platform collaborative scheduling,path following and motion control.This dissertation mainly focused on the technology of using inertial navigation to perform positioning and path following with self-avoidance function based on laser scanner.The architecture of the Mecanum wheeled vehicle studied in this paper was presented,and hardware selection and workflow designing were performed.The AGV positioning and path following techniques were studied in depth by combining multiple sources of information.It deduced the position estimation method of the Mecanum wheeled AGV odometry,and implemented a Mecanum vehicle localization algorithm using four encoders and a gyro in this dissertation.The simulation results shown that the algorithm has a significant effect on the accuracy of AGV positioning.The laser scanner technology including the measurement model and occupancy grid map creation was proposed,and an obstacle avoidance and a lane keeping method based on the laser scanner were designed.Experiments shown that AGV indoor navigation can be performed by using a laser scanner and an existing occupancy grid map.The last chapter presented the software implementation of the host computer,and several groups of experiments were designed to verify the system performance respectively.In the end,a Mecanum wheeled AGV path following process in a structured scene was performed.The result shown that the method produced in this paper was very effective and reliable.