The paper focuses on the risk-aware planning problem of Unmanned Ground Vehicles (UGVs) in off-road environments. It proposes a Coarse2fine A* global planning algorithm that incorporates a potential field approach to enhance safety while ensuring efficiency. A deterministic sampling method is also used for local planning, modified for off-road environments, and integrated with a risk assessment model for local risk avoidance. The algorithm’s performance was validated through simulation experiments and real-world system deployment, demonstrating its practicality and effectiveness.
Publication date: 6 Feb 2024
Project Page: unknown
Paper: https://arxiv.org/pdf/2402.02457
