Path Planning on Multi-level Point Cloud with a Weighted Traversability Graph

TL;DR

This work tackles 3D path planning for ground vehicles on multi-level terrains by integrating an efficient multi-level point-cloud map (ML-SkiMap) with a Weighted Traversability Graph (WTG) that encodes both surface connectivity and local safety. The method uses SkipList-based ML-SkiMap storage and a curvature-driven variable-resolution map to slim data while preserving geometry, then assigns multi-level, eight-direction traversability weights on a connectivity graph. A modified A* search on the WTG leverages these weights to compute short, safe paths, demonstrated through indoor/outdoor experiments and online datasets, including multi-floor structures. The approach offers robust, data-efficient 3D path planning for UGVs in complex environments, with practical applicability to scenarios involving occlusions, tip-over, and chassis collisions, while noting limitations with overhangs and potential outlier effects.

Abstract

This article proposes a new path planning method for addressing multi-level terrain situations. The proposed method includes innovations in three aspects: 1) the pre-processing of point cloud maps with a multi-level skip-list structure and data-slimming algorithm for well-organized and simplified map formalization and management, 2) the direct acquisition of local traversability indexes through vehicle and point cloud interaction analysis, which saves work in surface fitting, and 3) the assignment of traversability indexes on a multi-level connectivity graph to generate a weighted traversability graph for generally search-based path planning. The A* algorithm is modified to utilize the traversability graph to generate a short and safe path. The effectiveness and reliability of the proposed method are verified through indoor and outdoor experiments conducted in various environments, including multi-floor buildings, woodland, and rugged mountainous regions. The results demonstrate that the proposed method can properly address 3D path planning problems for ground vehicles in a wide range of situations.

Figures (29)

• Figure 1: (a) The test scene; (b) the original point cloud data set; (c) the curvature distribution map; (d) the various resolution map.
• Figure 2: Definition of the traversability vector $F_{i,j}$.
• Figure 3: Weighted traversability graph construction: (a) the local ML-SkiMap voxels; (b) the extraction of the local surface voxels around some $a_{i.j}$; (c) connectivity analysis of the multi-level surfaces around $a_{i.j}$; (d) traversability analysis at $a_{i.j}$ in 8 directions on two levels; (e) combining results of $C_{i,j}^l$ and $F_{i,j}^l$ to get the WTG.
• Figure 4: A simplified geometric model of the vehicle.
• Figure 5: Footprints of the vehicle and the four corresponding fitted planes to estimate the vehicle pose.