Enhancing Large-scale UAV Route Planing with Global and Local Features via Reinforcement Graph Fusion

TL;DR

The paper tackles UAV route planning under large-scale settings by reframing it as a large-scale Euclidean TSP (LSTSP) problem with up to $10{,}000$ sites. It introduces Delaunay Triangulation-based TSP Generalization Framework (DTTGF), a training-free, divide-and-conquer approach that preserves global structure through a Delaunay triangulated graph, solves subgraphs with embedded solvers, and fuses sub-solutions into a global heatmap. A pseudo-reinforcement warm-up strategy refines the fused heatmap to mitigate information loss from subgraph decomposition. Experiments on three large-scale datasets show that DTTGF enhances scalability and often matches or surpasses state-of-the-art methods without retraining, underscoring its practical impact for real-world UAVRP deployments.

Abstract

Numerous remarkable advancements have been made in accuracy, speed, and parallelism for solving the Unmanned Aerial Vehicle Route Planing (UAVRP). However, existing UAVRP solvers face challenges when attempting to scale effectively and efficiently for larger instances. In this paper, we present a generalization framework that enables current UAVRP solvers to robustly extend their capabilities to larger instances, accommodating up to 10,000 points, using widely recognized test sets. The UAVRP under a large number of patrol points is a typical large-scale TSP problem.Our proposed framework comprises three distinct steps. Firstly, we employ Delaunay triangulation to extract subgraphs from large instances while preserving global features. Secondly, we utilize an embedded TSP solver to obtain sub-results, followed by graph fusion. Finally, we implement a decoding strategy customizable to the user's requirements, resulting in high-quality solutions, complemented by a warming-up process for the heatmap. To demonstrate the flexibility of our approach, we integrate two representative TSP solvers into our framework and conduct a comprehensive comparative analysis against existing algorithms using large TSP benchmark datasets. The results unequivocally demonstrate that our framework efficiently scales existing TSP solvers to handle large instances and consistently outperforms state-of-the-art (SOTA) methods. Furthermore, since our proposed framework does not necessitate additional training or fine-tuning, we believe that its generality can significantly advance research on end-to-end UAVRP solvers, enabling the application of a broader range of methods to real-world scenarios.

Figures (1)

• Figure 1: Pipeline of the proposed approach. Step 1: perform Delaunay Triangulation(DT) on the current instance. Step 2: sampling and solving the sub-graph on DT results. Step 3: merging sub-solutions (one-stage solver) or sub-heatmaps (two-stage solver) to obtain a global heat map. Step 4: warming up heat map and search for the global optimal solutions.