TSPDiffuser: Diffusion Models as Learned Samplers for Traveling Salesperson Path Planning Problems

2024-06-05

arXiv.org

AI Summary

This work addresses the Traveling Salesman Problem with Path Planning (TSPPP) in obstacle-dense environments. We propose the first diffusion-model-based generative framework for TSPPP, modeling tours as conditional sequences that jointly encode obstacle maps and geometric path priors; collision-free, near-optimal paths are directly sampled via iterative denoising. Generated paths populate a lightweight path graph to implicitly estimate inter-destination traversal costs. Our key contribution lies in replacing conventional search or optimization paradigms with a generative modeling approach—enabling rapid sampling of high-quality tours and end-to-end cost estimation in a single unified framework. Evaluated across diverse indoor/outdoor real-world and synthetic scenarios, our method achieves significantly better trade-offs between solution quality and computational efficiency than state-of-the-art baselines, demonstrating superior performance across all metrics.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Planning, Routing, and Scheduling: Planning with Language Models
• Computer Vision: Diffusion Models for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

This paper presents TSPDiffuser, a novel data-driven path planner for traveling salesperson path planning problems (TSPPPs) in environments rich with obstacles. Given a set of destinations within obstacle maps, our objective is to efficiently find the shortest possible collision-free path that visits all the destinations. In TSPDiffuser, we train a diffusion model on a large collection of TSPPP instances and their respective solutions to generate plausible paths for unseen problem instances. The model can then be employed as a learned sampler to construct a roadmap that contains potential solutions with a small number of nodes and edges. This approach enables efficient and accurate estimation of travel costs between destinations, effectively addressing the primary computational challenge in solving TSPPPs. Experimental evaluations with diverse synthetic and real-world indoor/outdoor environments demonstrate the effectiveness of TSPDiffuser over existing methods in terms of the trade-off between solution quality and computational time requirements.

Problem

• Efficient path planning for TSPPPs
• Shortest collision-free path generation
• Diffusion model for TSPPP solutions

Innovation

• Diffusion models for path planning
• Learned sampler for TSP solutions
• Efficient collision-free path estimation