Motion planning involves decision making among combinatorial maneuver variants in urban driving. A planner must consider uncertainties and associated risks of the maneuver variants, and subsequently select a maneuver alternative. However, sometimes the uncertainty is so high that a reasonable decision is not possible. In such cases, the planner should postpones the combinatorial decision making to a later time and drive maneuver-neutral trajectory. In this way, safe but at the same time not overconservative motion can be planned.
The environment information is often quite noisy and has a tendency to contain false positive object detection. State-of-the-art motion planners consider all objects alike, thus producing overcautious behavior. My planning approach that considers alternative maneuvers in a combined fashion and plans a motion that is formed by the probabilities of those alternatives. The proposed planner can smoothly react to objects with low existence probability while remaining collision-free in case their existence substantiates.
Standard approaches calculate the homotopic maneuver alternatives m-A and m-B separately. Undecided motion of maneuver m-C obtained by jointly calculating m-A and m-B.
Combinatorial Reasoning in Motion Planning for Automated Vehicles
For my master's thesis at the department on Mobile Perception Systems of FZI Research Center for Information Technology, I focused on integrating combinatorial reasoning in continuous optimization methods. The results and insights allowed my supervisor and me to publish a paper on this topic, whose details are provided in the publication lists below. To gain more insight you can refer to my master's thesis or watch the video below.
Two trajectories of different homotopy classes in the 3-dimensional configuration space.