Discrete-Guided Diffusion for Scalable and Safe Multi-Robot Motion Planning

2025-08-27

AI Summary

Multi-robot motion planning (MRMP) faces a fundamental trade-off between scalability and trajectory quality: discrete multi-agent pathfinding (MAPF) scales well but yields coarse, piecewise-linear trajectories, while continuous optimization achieves high-quality smooth trajectories yet suffers from the curse of dimensionality. This paper proposes the Discrete-Guided Diffusion Framework (DGDF), the first approach to leverage discrete MAPF solutions as spatiotemporal guidance signals for a constrained generative diffusion model operating in convex configuration space to solve non-convex continuous trajectory optimization. A lightweight repair mechanism ensures collision-free execution. DGDF synergistically integrates the scalability of discrete search with the expressive power of generative modeling. Evaluated on high-density scenarios with up to 100 robots, it achieves efficient, high-quality, collision-free planning—demonstrating significantly higher success rates and computational efficiency than state-of-the-art methods. DGDF establishes a novel paradigm for large-scale MRMP.

Technology Category

• Planning, Routing, and Scheduling: Mixed Discrete/Continuous Planning
• Search and Optimization: Mixed Discrete/Continuous Search
• Intelligent Robots: Motion and Path Planning

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Multi-Robot Motion Planning (MRMP) involves generating collision-free trajectories for multiple robots operating in a shared continuous workspace. While discrete multi-agent path finding (MAPF) methods are broadly adopted due to their scalability, their coarse discretization severely limits trajectory quality. In contrast, continuous optimization-based planners offer higher-quality paths but suffer from the curse of dimensionality, resulting in poor scalability with respect to the number of robots. This paper tackles the limitations of these two approaches by introducing a novel framework that integrates discrete MAPF solvers with constrained generative diffusion models. The resulting framework, called Discrete-Guided Diffusion (DGD), has three key characteristics: (1) it decomposes the original nonconvex MRMP problem into tractable subproblems with convex configuration spaces, (2) it combines discrete MAPF solutions with constrained optimization techniques to guide diffusion models capture complex spatiotemporal dependencies among robots, and (3) it incorporates a lightweight constraint repair mechanism to ensure trajectory feasibility. The proposed method sets a new state-of-the-art performance in large-scale, complex environments, scaling to 100 robots while achieving planning efficiency and high success rates.

Problem

• Integrating discrete MAPF solvers with diffusion models for MRMP
• Overcoming scalability and trajectory quality limitations in multi-robot planning
• Ensuring collision-free trajectories in complex multi-robot environments

Innovation

• Combines discrete MAPF solvers with diffusion models
• Decomposes nonconvex MRMP into tractable convex subproblems
• Uses constrained optimization to guide diffusion models