群智感知场景下基于深度强化学习的多无人机路径规划算法
首发时间:2023-06-26
摘要:群智感知活动通常对需要感知的兴趣点要求较高的感知覆盖率,作为群智感知活动的重要参与者之一,无人机可以有效感知人类参与者无法感知和其他无人机还未感知的兴趣点。本文面向群智感知下的多无人机数据协同收集问题,基于深度强化学习理论模型提出了GMRL算法,并借助于大量模拟实验和小规模的真实实验证实本文所提供的算法的效率。
For information in English, please click here
A Deep Reinforcement Learning Algorithm for UAVs Route Planning in Mobile Crowd Sensing
Abstract:Mobile crowd sensing activities usually require a high coverage of the points of interest to be sensed. As one of the important participants, UAVs can effectively sense points of interest that are not sensed by human participants and not yet sensed by other UAVs. This paper proposes a GMRL algorithm based on deep reinforcement learning to solve the cooperative data collection problem of multiple UAVs. The efficiency of this algorithm is confirmed with the help of a large number of simulations and small-scale real experiments.
Keywords: Deep reinforcement learning Multi-agent system Mobile crowd sensing Route planning
引用
No.****
动态公开评议
共计0人参与
勘误表
群智感知场景下基于深度强化学习的多无人机路径规划算法
评论
全部评论