Determining forbidden steering directions for a passenger car in urban environments based on the velocity obstacle approach and use of trackers

One of the most challenging tasks for autonomous passenger cars in urban-like environments is to select the ideal steering angle to avoid collisions with pedestrians and other vehicles. In order to determine the set of steering angles that must be avoided it is necessary to track mobile obstacles that surround the vehicle. This task is especially difficult in urban environments where a great variety of obstacles may induce the vehicle embedded controllers to take erroneous decisions. They need as much information as possible concerning the obstacle positions and speeds (direction and magnitude) in order to plan evasive maneuvers that avoid collisions. Unfortunately, obstacles close to vehicle embedded sensors frequently cause blind zones behind them where other obstacles could be hidden. In order to overcome this problem we developed an obstacle tracking module based only on 2D laser scan data. Its main parts consist of obstacle detection, obstacle classification, and obstacle tracking. In addition to this, we implemented a modified version of the velocity obstacle approach to determine the set of forbidden steering angles for a certain time window. The real data samplings were acquired in an urban-like environment in our University Campus using our test vehicle. The tests proved the applicability of the algorithms in urban-like environments.