Efficient Spatiotemporal Graph Search for Local Trajectory Planning on Oval Race Tracks

Autonomous racing has increasingly become a research subject as it provides insights into dynamic, high-speed situations. One crucial aspect of handling these situations, especially in the presence dynamic obstacles, is generation collision-free trajectory that represents safe behavior and also competitive case racing. We propose local planning approach generates such trajectories for car on an oval race track by searching spatiotemporal graph. A considerable challenge search-based methods domain curse dimensionality. Therefore, we how previously presented graph structure based intervals instead discrete values can be searched more efficiently without losing optimality using uniform-cost search strategy. extend method to make anytime-capable so provide suboptimal even if terminated early. The graph-based allows us apply flexible cost function our operate fully autonomously track, including pit lane. present explain terms contribute desired behaviors. This supported results with full-scale prototype.