Pedestrians Walking on Reachable Sets and Manifolds

It is a challenge to find safe trajectories for automated vehicles. Especially in urban environments with pedestrians there exist many different situations. The prediction of future movements with absolute certainty is impossible if the intention of the pedestrian is unknown. This paper is intended to offer a new concept of decision-making for motion planning with vulnerable road-users in urban environments. Reachability analysis is used to define spatial areas based on physical constraints, where the pedestrian might go. The approach is physically inspired and it can be proven for all kind of pedestrians. A new approach to cluster spatial areas is introduced to store specific environmental information about the pedestrian. The maximum values for jerk, acceleration and velocity of the pedestrian are necessary to find reachable areas.