The article discusses the issue of path planning for multiple non-holonomic robots in continuous domains, a significant challenge in robotics. The authors propose two algorithms designed to solve static/one-shot and lifelong path/motion planning tasks for Ackerman car-like robots. The first algorithm, Priority-inherited Backtracking for Car-like Robots (PBCR), adapts a decentralized strategy. The second algorithm, Enhanced Conflict-based search for Car-like Robots (ECCR), is a centralized method. The methods are evaluated through simulation and real-world experiments, confirming their effectiveness.
Publication date: 20 Feb 2024
Project Page: Not provided
Paper: https://arxiv.org/pdf/2402.11767
