A Control Method to Avoid Obstacles for an Intelligent Car Based on Rough Sets and Neighborhood Systems

Abstract

In order to study the dynamic control model of intelligent cars following roads with obstacles, this paper presents a control method to avoid obstacles for an intelligent car. Firstly, to simplify the shape of an obstacle, according to the related theory of covering rough set model, we use a finite number of trapezoid neighborhoods to cover the irregular obstacle. And whose upper approximation in the sense of the rough set is used as a regular obstacle region. Secondly, an orientation method of intelligent cars' relative to roads with regular obstacles based on feasible neighborhoods is presented. And the feasible neighborhood of a certain moment is determined by the lower approximation of the detected feasible neighborhood, which is covered by a finite number of standard trapezoid neighborhoods and regarded as a rough set in the sense of covering rough set model. The effectiveness of the presented methods is illustrated by simulation experiments with Matlab.