Moving obstacle detection algorithm using cell decomposition method

In this paper, the algorithm to detect obstacles surrounding an autonomous vehicle and the method to navigate this vehicle on the road are studied. For this purpose, the road is divided into cells in lateral and longitudinal directions. The assumption is that some special tools specify the cells positions and then full and empty-cell corresponding matrix is generated. In this matrix, full cells are displayed with digit 1 and empty cells are displayed with digit 0. In the next step, by analyzing the matrix in Matlab, the vehicle is navigated. In this analysis, firstly the position of the vehicle and the obstacles are identified. Then, based on the road conditions and the obstacles positions, required orders to move the vehicle are determined. If a lane change is needed, according to the road’s curvature and the distance between the vehicle and the obstacle, appropriate path for the vehicle will be chosen. In this paper, for the first time in autonomous vehicle navigations, the road is considered as a 1 and 0 matrix. In this method, the road matrix gets updated with time and provides the possibility of analyzing the vehicle’s movement. Also, the algorithm used to solve the problem is very simple.