An Obstacle Detection Algorithm Based on Ultrasonic Sensors for Autonomous Land Vehicle

An dynamic filtering based obstacle detection algorithm is proposed for the navigation and control of Autonomous Land Vehicle (ALV). The algorithm detects the obstacles using sequential sonar data from dual sonar sensors. The sonar model is described at first. Then the obstacles' features, depicted as lines, are extracted. Finally a dynamic data filtering algorithm is described, in which the sonar return data is firstly processed through dynamic filtering using the orientation and the trajectory information of the vehicle. The algorithm's validity is approved through a field test in cross-country environment.