기지 환경 하의 전방향구동용 무인지상차량을 위한 다중목표점 경로계획 및 추적제어기 설계

Abstract

Today, Automatic Guided Vehicles (AGVs) with a path planning are used in many industrial fields. Mainly, Automatic Guided Vehicle (AGV) is a convenient mobile robot that cleans a room or transports industrial material in a known environment. In order to be used more safely and widely, AGV needs to be able to plan a path for itself and furthermore provide a tracking controller. However, the existing path planning algorithms have the disadvantage of traversing between obstacles or passing over the sharp edges of obstacles. In order to solve these weaknesses, a new path algorithm to set multiple target points in the existing path planning and a controller design to track the planned paths are needed deeply. This thesis proposes multiple target point path planning for an Omnidirectional Automatic Guided Vehicle (OAGV) and designs a tracking controller to track paths generated by the proposed path planning. The path planning algorithms include various algorithms such as A*, D*, D* Lite. These three path planning algorithms are focused on moving from the starting point to the target point in the shortest path avoiding obstacles in a known environment. Multiple target points are created with a modified D* Lite algorithm based on the most efficient D* Lite algorithm among these three algorithms. To do this task, the followings are done. Firstly, it is assumed that the workspace is divided into cells and the size of one cell is the size of OAGV. Secondly, a method of setting multiple target points using the Hamiltonian path is presented. Thirdly, the configuration of the OAGV used in the experiment is described. The OAGV system consists of several interconnected devices such as main controller, monitor, keyboard, mouse, DC servo driver, and industrial computer, the omnidirectional drive wheel, and so on. The laser sensor NAV-200 is used to detect the OAGV posture in known environment. Fourthly, kinematic modeling and dynamic modeling of the OAGV are presented. In particular, a tracking controller that tracks paths obtained by the multiple target point path planning algorithm is presented. A tracking controller is designed using the backstepping method, and the stability of the designed tracking controller is proven by the Lyapunov stability theory. Finally, simulation shows the comparison of the multiple target point path planning and D* Lite algorithm. Simulation and Experimental results are shown to verify the effectiveness and the performance of the proposed path planning algorithm and the proposed controller.