Distributed multi-UAV cooperation in UAV-assisted mobile crowd sensing based on information exchange

作者：Guisong Yang, Mengdie Liu, Xing He · 发表于：Advances in Engineering Innovation · 年份：2026 · DOI：10.54254/2977-3903/2026.32627 · 研究领域：Mobile Crowdsensing and Crowdsourcing、UAV Applications and Optimization、Distributed Control Multi-Agent Systems

In UAV-assisted Mobile Crowd Sensing, ground workers are responsible for basic data collection, while UAVs can perform supplementary sensing in areas that are difficult to cover in a timely manner. To address the limited efficiency of distributed multi-UAV cooperative supplementary sensing in the absence of a central controller, this paper proposes an information-exchange-based distributed multi-UAV cooperative sensing algorithm, termed IE-DDQN. The proposed method formulates the problem as a decentralized partially observable Markov decision process, enables inter-UAV information exchange through short-range Device-to-Device (D2D) links, and employs a neighbor information pooling module to aggregate multi-source high-dimensional messages into low-dimensional representations, which are then combined with local observations for decision making. Experimental results demonstrate that the proposed method outperforms pure-worker and pure-UAV schemes in terms of weighted data value per unit composite cost, task completion rate, and task completion time.