Scholarlink

In this paper, we investigate a DRL-based method for UAV trajectory design and mobile ground user (GU) scheduling in a UAV-assisted mobile edge computing (MEC) system. In this scenario, the UAV first collects data from source GUs, processes it, and then transmits the processed data to the destination GUs. Our objective is to minimize the total operation time, which includes data collection, computation, and decision transmission, by optimizing the UAV’s three-dimensional (3D) trajectory and mobile GUs scheduling. In this scenario, considering the mobility of users and obstacle avoidance with 3D buildings, the problem becomes non-convex, and the environment is dynamically changing, making it difficult for traditional methods to address. Therefore, we propose a DRL-based approach for UAV trajectory design and user scheduling. This method only allows the UAV to interact with the environment multiple times, avoiding the challenges of non-convexity. The simulation results demonstrate the superiority of the proposed algorithm. Compared to 2D trajectory design, the 3D trajectory performs better due to the utilization of vertical altitude. Additionally, the simulations confirm the algorithm’s robustness across different user mobility patterns.

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

DRL-Based Joint User Scheduling and UAV Trajectory Design in UAV-Assisted MEC System

  • 章节

  • 发表日期：

    2025年

查看全文
  • Yunfei Gao,
  • Xiaopeng Yuan,
  • Qinwei He,
  • Yulin Hu

摘要

In this paper, we investigate a DRL-based method for UAV trajectory design and mobile ground user (GU) scheduling in a UAV-assisted mobile edge computing (MEC) system. In this scenario, the UAV first collects data from source GUs, processes it, and then transmits the processed data to the destination GUs. Our objective is to minimize the total operation time, which includes data collection, computation, and decision transmission, by optimizing the UAV’s three-dimensional (3D) trajectory and mobile GUs scheduling. In this scenario, considering the mobility of users and obstacle avoidance with 3D buildings, the problem becomes non-convex, and the environment is dynamically changing, making it difficult for traditional methods to address. Therefore, we propose a DRL-based approach for UAV trajectory design and user scheduling. This method only allows the UAV to interact with the environment multiple times, avoiding the challenges of non-convexity. The simulation results demonstrate the superiority of the proposed algorithm. Compared to 2D trajectory design, the 3D trajectory performs better due to the utilization of vertical altitude. Additionally, the simulations confirm the algorithm’s robustness across different user mobility patterns.