Problem
Research questions and friction points this paper is trying to address.
Multi-Agent Path Finding
Generalised MAPF
Continuous-time Planning
Heterogeneous Agents
Non-geometric Conflicts
AOC-CBS: Anytime-Optimal Continuous-time Conflict-Based Search for Generalised Multi-Agent Path Finding
🤖 AI Summary
Traditional multi-agent path finding (MAPF) is limited by assumptions of discrete time, circular agents, and single-task scenarios, rendering it inadequate for real-world challenges such as heterogeneous agents, non-geometric conflicts, and complex task sequences. This work proposes a generalized MAPF framework and introduces the Anytime-Optimal Continuous-time Conflict-Based Search (AOC-CBS) solver, which supports arbitrary agent shapes, dynamically feasible trajectories, and multi-stage tasks in continuous time. AOC-CBS achieves, for the first time, an exact, resolution-complete, and anytime-optimal solution to generalized MAPF. It incorporates a Tier-Prioritized Safe Interval Path Planning repair mechanism, integrating multi-strategy conflict resolution, continuous-time collision detection, and multi-threaded parallel optimization. Experiments demonstrate that AOC-CBS significantly improves scalability in scenarios with hundreds of non-convex, heterogeneous agents, outperforming the OC-CBS baseline under controllable optimality bounds while preserving solution optimality.
📝 Abstract
Many research fields share a common structure: a set of agents, each pursuing its own goal, whose actions must be coordinated so that no two of them conflict. Multi-Agent Path Finding (MAPF) is a concrete instance of this structure, with applications from warehouses to road traffic and airports. Much of MAPF research assumes discrete time, circular agents sharing one spatial graph, a single goal per agent, and that an agent must remain at its goal once reached, precluding heterogeneous fleets, non-geometric conflicts, task sequences, and agents that move on after completing them. We generalise the MAPF formulation to lift these assumptions, and present Anytime-Optimal Continuous-time Conflict-Based Search (AOC-CBS), an exact and solution-complete solver for it. AOC-CBS guarantees the eventual return of an optimal solution, while reporting an incumbent with a known optimality gap upper bound throughout its runtime; it is configurable with a portfolio of repair functions, one of which we introduce (Tier-Prioritized Safe Interval Path Planning), and can exploit multiple processor cores. We demonstrate AOC-CBS on a mixed fleet of non-convex agents moving along smooth, kinodynamically feasible trajectories. Preliminary experiments against the exact solver OC-CBS, on well-known benchmarks and roadmaps we sample from them, show AOC-CBS is comparable at finding optimal solutions while extending scalability from the tens to the hundreds of agents when a bounded optimality gap is accepted.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Anytime-optimal
Continuous-time MAPF
Conflict-Based Search
Heterogeneous agents
Kinodynamic trajectories
🔎 Similar Papers
2024-10-02arXiv.orgCitations: 0
