Academic Project Page

Contiguity-Driven Task Allocation with Connectivity-Aware Task Representation for Decentralized Multi-UAV Exploration

Xinlu Yan1,*, Mingjie Zhang2,3,*, Yuhao Fang1, Yanke Sun1,
Jun Ma2, Youmin Gong1, Boyu Zhou3,†, Jie Mei1,†
1 Harbin Institute of Technology, Shenzhen
2 The Hong Kong University of Science and Technology (Guangzhou)
3 Southern University of Science and Technology
* Equal Contribution    † Co-corresponding Authors
arXiv Code Bilibili Video

We propose C2-Explorer, a decentralized multi-UAV exploration framework that enhances the flexibility and contiguity of task allocation strategy. By integrating connectivity-aware task representation with contiguity-driven allocation, it decomposes disconnected unknown regions into independent task units and enables more efficient, communication-limited multi-UAV exploration with reduced cross-region detours.

Abstract

Efficient multi-UAV exploration under limited communication is severely bottlenecked by inadequate task representation and allocation. Previous task representations either impose heavy communication requirements for coordination or lack the flexibility to handle complex environments, often leading to inefficient traversal. Furthermore, short-horizon allocation strategies neglect spatiotemporal contiguity, causing non-contiguous assignments and frequent cross-region detours. To address this, we propose C2-Explorer, a decentralized framework that constructs a connectivity graph to decompose disconnected unknown components into independent task units. We then introduce a contiguity-driven allocation formulation with a graph-based neighborhood penalty to discourage non-adjacent assignments, promoting more contiguous task sequences over time. Extensive simulation experiments show that C2-Explorer consistently outperforms state-of-the-art (SOTA) baselines, reducing average exploration time by 43.1% and path length by 33.3%. Real-world flights further demonstrate the system's feasibility.

Motivations & Contributions

Motivations

Inadequate Task Representation click

Topology-agnostic representation overlooks the underlying connectivity of the regions.

Non-Contiguous Task Assignment click

Existing allocation strategies often fail to consider spatiotemporal contiguity.

Contributions

01

Connectivity-Aware Task Representation click

We construct a connectivity graph that decomposes disconnected unknown regions into independent task units, enabling flexible and communication-efficient task management.

02

Contiguity-Driven Task Allocation click

We introduce a contiguity-driven allocation formulation with a graph-based neighborhood penalty to discourage non-adjacent task assignments, promoting spatiotemporally contiguous allocation sequences over time.

03

State-of-the-Art Performance click

C2-Explorer reduces average exploration time by 43.1% and path length by 33.3% compared to SOTA baselines, with real-world flights validating the system's feasibility.

Benchmark Comparison

4 Drones 8× speed

C2-Explorer

RACER

FAME

Method Exploration Time (s) Total Path Length (m) Flight Vel. (m/s)

bold = best  ·  underline = 2nd best

BibTeX

@misc{yan2026c2explorer,
      title={C$^2$-Explorer: Contiguity-Driven Task Allocation with Connectivity-Aware Task Representation for Decentralized Multi-UAV Exploration}, 
      author={Xinlu Yan and Mingjie Zhang and Yuhao Fang and Yanke Sun and Jun Ma and Youmin Gong and Boyu Zhou and Jie Mei},
      year={2026},
      eprint={2603.07699},
      archivePrefix={arXiv},
      primaryClass={cs.RO},
      url={https://arxiv.org/abs/2603.07699}, 
}