# Fast and Communication-Efficient Multi-UAV Exploration Via Voronoi Partition on Dynamic Topological Graph

**Authors:** Qianli Dong, Haobo Xi, Shiyong Zhang, Qingchen Bi, Tianyi Li, Ziyu Wang, Xuebo Zhang  
**arXiv:** [2408.05808](https://arxiv.org/abs/2408.05808)  
**Date:** 2024-08-11  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/2408.05808v1/

## Key points

- Novel approach to multi-UAV exploration in large environments
- Introduction of multi-robot dynamic topological graph (MR-DTG) to reduce communication volume
- Hierarchical multi-UAV exploration framework using graph Voronoi partition for task allocation
- Validation through extensive simulation and real-world experiments showing reduced exploration time and communication volume
- First use of graph Voronoi partition for multi-UAV exploration, demonstrating effectiveness
- Comprehensive review of related works in autonomous exploration and discussion on limitations
- Promising solution to enhance multi-UAV exploration efficiency through innovative techniques and optimized strategies
- Release of source code for replication and extension of study's findings

## Summary

In this paper, the authors present a novel approach to multi-UAV exploration in large environments. The focus is on improving efficiency through efficient data transmission and task allocation. The key contribution is the introduction of a multi-robot dynamic topological graph (MR-DTG) that reduces communication volume by transferring only necessary information for exploration planning. This serves as the foundation for a hierarchical multi-UAV exploration framework utilizing graph Voronoi partition for reasonable task allocation. Extensive simulation and real-world experiments validate the proposed method against state-of-the-art techniques, showing significant improvements in terms of reduced exploration time and communication volume. Notably, this work is the first to use graph Voronoi partition for multi-UAV exploration, showcasing its effectiveness. The paper also provides a comprehensive review of related works in autonomous exploration and discusses existing methods' limitations in scalability and computational complexity. Overall, the proposed method offers a promising solution to enhance multi-UAV exploration efficiency through innovative techniques and optimized strategies. The release of source code further contributes to the research community's benefit by enabling replication and extension of the study's findings.