SphereMap: Dynamic Multi-Layer Graph Structure for Rapid Safety-Aware UAV Planning

TL;DR

SphereMap introduces a dynamic, multi-layer graph structure for UAV planning that enhances safety and speed, enabling efficient real-time navigation and multi-robot cooperation in complex environments.

Contribution

The paper presents a novel two-layer graph representation built with intersecting spheres for rapid, safety-aware UAV planning and decentralized environment mapping.

Findings

Achieved faster planning than state-of-the-art methods.

Maintained safer distances from obstacles during navigation.

Secured second place in DARPA SubT Challenge virtual track.

Abstract

A flexible topological representation consisting of a two-layer graph structure built on-board an Unmanned Aerial Vehicle (UAV) by continuously filling the free space of an occupancy map with intersecting spheres is proposed in this \paper{}. Most state-of-the-art planning methods find the shortest paths while keeping the UAV at a pre-defined distance from obstacles. Planning over the proposed structure reaches this pre-defined distance only when necessary, maintaining a safer distance otherwise, while also being orders of magnitude faster than other state-of-the-art methods. Furthermore, we demonstrate how this graph representation can be converted into a lightweight shareable topological-volumetric map of the environment, which enables decentralized multi-robot cooperation. The proposed approach was successfully validated in several kilometers of real subterranean environments, such as caves, devastated industrial buildings, and in the harsh and complex setting of the final event of the DARPA SubT Challenge, which aims to mimic the conditions of real search and rescue missions as closely as possible, and where our approach achieved the \nth{2} place in the virtual track.