An odometry free automatic perpendicular parking strategy for a light urban vehicle based on a low resolution lidar

In this paper, an automatic perpendicular parking algorithm for a light urban vehicle with active steer assembly is presented, designed to help the user when the manoeuvre has to be performed in a narrow space between two obstacles. The parking strategy is odometry free and based only on the measurements from a limited FOV low resolution Lidar, resulting in a simple and cost effective solution. In the initial phase of the manoeuvre, the algorithm exploits the Lidar information - combined with the a priori knowledge of the park structure - to localize the vehicle in the parking environment and to generate the steering action accordingly, so as to approach the parking space. Then, as the vehicle approaches the parking spot entrance, an obstacle avoidance logic is applied to end the manoeuvre without crashes. The performance of the proposed strategy has been assessed through an extensive simulation campaign and validated by experimental tests using a real instrumented vehicle.