A Skeleton-Based Topological Planner for Exploration in Complex Unknown Environments

TL;DR

This work tackles efficient autonomous exploration in unknown environments by addressing back-and-forth maneuvers and computational burden inherent in greedy and globally optimized methods. It introduces STGPlanner, a skeleton-based topological planner that builds a skeletal topological graph (STG) from incremental skeleton extraction and uses a finite state machine to drive exploration without heavy TSP or ray-casting computations. The main contributions are an incremental skeleton extraction method based on wavefront propagation, a skeletal topological graph update with expansion and simplification, and an STG-based FSM exploration strategy that leverages topology to achieve fast, low-overhead exploration. Experimental results in four simulated scenes and a real-world indoor test demonstrate improved exploration efficiency and substantially reduced computation compared with state-of-the-art methods, highlighting the practical impact of topology-guided exploration for robotic systems.

Abstract

The capability of autonomous exploration in complex, unknown environments is important in many robotic applications. While recent research on autonomous exploration have achieved much progress, there are still limitations, e.g., existing methods relying on greedy heuristics or optimal path planning are often hindered by repetitive paths and high computational demands. To address such limitations, we propose a novel exploration framework that utilizes the global topology information of observed environment to improve exploration efficiency while reducing computational overhead. Specifically, global information is utilized based on a skeletal topological graph representation of the environment geometry. We first propose an incremental skeleton extraction method based on wavefront propagation, based on which we then design an approach to generate a lightweight topological graph that can effectively capture the environment's structural characteristics. Building upon this, we introduce a finite state machine that leverages the topological structure to efficiently plan coverage paths, which can substantially mitigate the back-and-forth maneuvers (BFMs) problem. Experimental results demonstrate the superiority of our method in comparison with state-of-the-art methods. The source code will be made publicly available at: https://github.com/Haochen-Niu/STGPlanner.

Figures (7)

• Figure 1: An overview of the proposed STGPlanner.
• Figure 2: Illustration of the skeleton extraction and STG update. (a) and (b) depict the wavefront propagation from unknown and occupied area at time $T_i$. As robot moves from $T_i$ to $T_{i+1}$ in (c), wavefront from occupied is preserved (shaded), thereby requiring only partial data processing. (d) provides an overview of the STG.
• Figure 3: Illustration of the exploration strategy FSM.
• Figure 4: The four simulation environments.
• Figure 5: The explored area over time. The shaded areas show the upper and lower bounds.