Efficient Path Planning in Large Unknown Environments with Switchable System Models for Automated Vehicles

TL;DR

This work tackles efficient path planning for large, largely unknown environments by introducing an early stopping navigation strategy that halts planning when a sufficient progress threshold is reached, and a divergence-based replanning scheme that triggers when two successive 2D A*-paths significantly diverge. It also extends Hybrid A* with switchable system models and rear-axis rotation capabilities to enable concise maneuvers for the U-Shift vehicle in narrow spaces. Empirical results show the guided planning approach substantially reduces runtime (up to ~$2\times$ cumulative and ~$4.6\times$ average) while preserving path smoothness and clearance, and the extended planner improves maneuverability and path conciseness in tight corridors. Together, these contributions enable real-time planning in large unknown environments and practical deployment for vehicles with advanced maneuvering capabilities like the U-Shift II.

Abstract

Large environments are challenging for path planning algorithms as the size of the configuration space increases. Furthermore, if the environment is mainly unexplored, large amounts of the path are planned through unknown areas. Hence, a complete replanning of the entire path occurs whenever the path collides with newly discovered obstacles. We propose a novel method that stops the path planning algorithm after a certain distance. It is used to navigate the algorithm in large environments and is not prone to problems of existing navigation approaches. Furthermore, we developed a method to detect significant environment changes to allow a more efficient replanning. At last, we extend the path planner to be used in the U-Shift concept vehicle. It can switch to another system model and rotate around the center of its rear axis. The results show that the proposed methods generate nearly identical paths compared to the standard Hybrid A* while drastically reducing the execution time. Furthermore, we show that the extended path planning algorithm enables the efficient use of the maneuvering capabilities of the concept vehicle to plan concise paths in narrow environments.

Figures (9)

• Figure 1: Contributions of this paper: a) The early stopping method stops the path planning algorithm to plan paths more efficiently. b) The replanning scheme defines the start pose. c) A path planning algorithm that switches between two system models to use the maneuvering capabilities of the U-Shift II concept vehicle ushift1ushift2.
• Figure 2: Schematic visualization of the early stopping method: The start node is shown as a blue cross, while the goal node is shown as a red one. In the gray area, the distance heuristic decreased for at least $s\textsubscript{w}$.
• Figure 3: Detection of diverging A*-paths. The first deviation of the two paths is not significant and is therefore ignored. The second deviation, marked by the red circle, is significant.
• Figure 4: a): The extended motion primitives are shown in blue along with the standard ones in red. b): Geometric extension: The intersection of the line segment of the current node's pose $P\textsubscript{c}$ (green) with the line segment of the goal's pose $P\textsubscript{g}$ (red) defines the point of rotation. The path to drive is shown in blue.
• Figure 5: Planned paths of the research vehicle of Ulm University (blue) to the goal pose (red) with the explored graph. a): Guidance by waypoint navigation. b): Guidance by early stopping.