Obstacle Avoidance And Path Planning Of Autonomous Ground Vehicle

Autonomous Ground Vehicles have many valuable attributes that can benefit human beings in all fields of modern life.Several important AGV technologies are discussed in this paper, include: Path Planning, Obstacle Avoidance, Perception Technologies; Control System Architecture and etc. All the research work in this paper are not only discussed theoretically but also performed with the intelligent four-wheeled vehicle"Explorer". The main content and achievements are as follows:The ideal kinematic model of the"Explorer"is introduced, based on which the localization method—Dead Reckoning algorithm is discussed carefully, and the error compensation methods are presented.The perception technologies is discussed which includes the sensors and the deployment scheme. With deliberative considerations, the Explorer's perception subsystem is established which is consist of one sonar sensor and several red-infrared sensors.Path planning is the most important module for autonomous ability. In this paper the Explorer's path planner has two mode: the obstacle-avoidance mode and the non-obstacle path planner mode, which are all achieved by fuzzy controllers.The optimized fuzzy control algorithm SIA (Storage-Inference Algorithm) is provided and proved can spare a large number of computation compared with common algorithm.The path planning puzzle ("U shape trap") has been related and solved theoretically in this paper.The Hybrid Deliberative-Reactive Architecture is introduced, on which the Explorer's control system is based. In terms of the system architecture, two-CPU structured hardware is designed as well.Finally, the experiments'results of the Explorer under different conditions are presented.