基于通信任务的无人机路径规划方法
首发时间:2021-04-26
摘要:随着无人机及其相关技术的日渐成熟,无人机作为空中基站被广泛应用于通信领域为地面用户提供通信服务。在诸多通信场景中,具有高机动性和自由性的无人机基站,可根据通信需求调整自身位置,通过移动提高通信质量,为无线通信场景提供辅助和扩展,研究通信场景中的无人机路径规划问题,对促进无线通信网络集成具有重要意义。本文首先分析了现有路径规划方法的优势和不足,其次根据通信场景中任务的需求,选择了基于深度强化学习的路径规划方法进行分析,最后针对深度强化学习方法的动作粗粒度问题,引入了行动者-批评家(Actor-Critic,AC)框架,将离散动作空间转化为连续动作空间,细化动作粒度,增大动作空间使得解空间增大,更逼近最优路径解。仿真结果表明,本文提出的优化方法效果优于基准方法。
关键词: 网络技术与应用 通信任务 路径规划 行动者-评论家框架?
For information in English, please click here
Path planning method for Unmanned Aerial Vehicle based on communication task
Abstract:With the development of UAV and its related technologies, UAV as a base station in the air has been widely used in the field of communication to provide communication services for ground users. In many communication scenario, unmanned aerial vehicle (UAV) with high mobility and freedom base, its position can be adjusted according to communication demand, through the mobile communication quality, provides the auxiliary and extension for wireless communication scenario, the UAV path planning problem of the research on communication scenarios, is of great significance to promote integrated wireless communication network. Firstly, this paper analyzes the advantages and disadvantages of the existing path planning methods. Secondly, the path planning method based on deep reinforcement learning is selected for analysis according to the requirements of tasks in communication scenarios. Finally, for the coarse-grained problem of deep reinforcement learning, actor-critic (AC) framework is introduced to transform the discrete action space into a continuous action space, refine the action granularity, and increase the action space to enlarge the solution space, which is closer to the optimal path solution. Simulation results show that the effectiveness of the proposed optimization method is better than that of the benchmark method.
Keywords: Network technology and application Communication Task Path Planning Actor-critic
基金:
引用
No.****
动态公开评议
共计0人参与
勘误表
基于通信任务的无人机路径规划方法
评论
全部评论0/1000