Autonomous mobile robot navigation using fuzzy logic control

摘要

Traditionally the type of robot used in the workplace consisted mainly o f the fixed arm variety. Any mobile robots that were commercially available required that the environment be altered to accommodate them. This involved the installation of guide lanes or some form of sensor units placed at various locations around the workplace to facilitate the robot in determining its position within the environment.ududSuch approaches are costly and limit the use of robots to environments where these methods are feasible. The inadequacies in this technology has led to research into autonomous mobile robots that offer greater flexibility and do not require changes in the enviromnent. There are many technical issues to be addressed in designing such a robot. These stem from the necessity that the robot must be able to navigate through an environment unaided. Other problems such as the cost of the vehicle must be considered so that prospective customers will not be put off.ududThis thesis discusses the strategies taken in addressing the problems associated with navigation in an obstacle strewn environment. Such issues include position estimation, path planning, obstacle avoidance and the acquisition and interpretation of sensor information. It also discusses the suitability of fuzzy logic for controlling a robot.ududA graphical user interface runs on the PC which communicates with the robot over a radio link. The robot uses a fuzzy logic controller to follow a planned path and avoid unknown obstacles by controlling the velocity and steering angle o f the drive unit. It is a tracked vehicle which is suitable for indoor use only. The results of path planning and the robots attempts at following the paths and avoiding obstacles are illustrated and discussed.