Study On Obstacle Avoidance For Wheeled Mobile Robot

To realize the autonomous obstacle avoidance of mobile robots, a series of methods are presented in this paper, which include circumstance detection around a mobile robot by ultrasonic sensors, obstacle input data process through microcomputer techniques and pattern recognition of obstacles with a fuzzy-neuro control algorithm. It mainly involves the following works:Circumstance data acquisition is accomplished by 8 US series BTE054 ultrasonic sensors , and circumstance input data process is completed through a microcomputer embedded with a Japanese Hitachi SH7044(QFP-112Pin)chip. In order to increase the measurement precision of single ultrasonic sensor and decrease the interference degree produced by multiple ultrasonic sensors, this paper provides a hardware method called multi-connection, and two software ways named median and intelligent forecast. This paper also presents a novel pattern recognition algorithm that is used to achieve autonomous obstacle avoidance of a mobile robot. Through the combination of heuristic fuzzy rules and the Kohonen clustering network, a heuristic fuzzy-neuro network is developed. This method also builds up pattern-mapping between ultrasonic sensory input data and velocity command by applying the off-line and unsupervised training method to this network, allowing continuous, fast motion for obstacle avoidance.In order to verify the effectiveness of these proposed methods, this paper finally gives the results of all kinds of obstacle avoidance simulation in a computer virtual environment with Visual C++ and also presents wealthy real experiments indoors.