Scholarlink

The utilization of unmanned aerial vehicles (UAVs) for data collection has become a prominent focus in the field of three-dimensional coverage in wireless sensor networks. However, the energy consumption during UAV flight and hovering remains a significant bottleneck affecting data collection efficiency. To address this issue, this paper presents a novel energy consumption model that accurately evaluates the energy usage of UAVs during both flight and hovering phases. Furthermore, a Q-learning-based Backtracking Search Algorithm (QL-BSA) is proposed to determine the minimal set of hover points and their optimal locations. In the QL-BSA framework, the sole control parameter of the Backtracking Search Algorithm is dynamically updated through continuous interaction between the agent and the environment, thereby enhancing the algorithm’s search performance. Experimental results demonstrate that the proposed approach outperforms existing state-of-the-art methods in terms of both energy efficiency and coverage optimization.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Energy Consumption Optimization of UAV Data Acquisition Based on Reinforcement Learning

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Li Tan,
  • Haixia Zhao,
  • Yuzhao Liu

摘要

The utilization of unmanned aerial vehicles (UAVs) for data collection has become a prominent focus in the field of three-dimensional coverage in wireless sensor networks. However, the energy consumption during UAV flight and hovering remains a significant bottleneck affecting data collection efficiency. To address this issue, this paper presents a novel energy consumption model that accurately evaluates the energy usage of UAVs during both flight and hovering phases. Furthermore, a Q-learning-based Backtracking Search Algorithm (QL-BSA) is proposed to determine the minimal set of hover points and their optimal locations. In the QL-BSA framework, the sole control parameter of the Backtracking Search Algorithm is dynamically updated through continuous interaction between the agent and the environment, thereby enhancing the algorithm’s search performance. Experimental results demonstrate that the proposed approach outperforms existing state-of-the-art methods in terms of both energy efficiency and coverage optimization.