工程科学学报2025，Vol.47Issue(1)：91-100,10.DOI:10.13374/j.issn2095-9389.2024.06.01.001
面向工业场景的无人机时空众包资源分配
UAV spatiotemporal crowdsourcing resource allocation based on deep reinforcement learning
摘要
Abstract
Spatiotemporal crowdsourcing involves the use of various Internet of Things(IoT)devices distributed across industrial environments to collect and transmit spatiotemporal data related to industrial operations.Unmanned aerial vehicles(UAVs)play a crucial role in further collecting this data from IoT devices,especially in spatiotemporal crowdsourcing tasks.In the realm of industrial IoT energy management,allocating spatiotemporal crowdsourcing resources to UAVs poses substantial challenges.Traditional approaches to this problem have focused on optimizing the Age of Information(AoI)to ensure timely and equitable data updates.However,these methods often overlook critical operational constraints such as UAV no-fly zones and the risk of data interception by eavesdroppers.These issues can adversely affect the freshness and integrity of the information being gathered and transmitted.To address these shortcomings,this paper presents a novel deep reinforcement learning-based framework for UAV spatiotemporal crowdsourcing resource allocation.This approach aims to minimize the average AoI across the network while also reducing the energy consumption of IoT devices.It incorporates spatial constraints imposed by UAV no-fly zones and actively manages the transmission of jamming signals to mitigate the threat posed by eavesdroppers,thus ensuring data security.However,the complexity of allocating spatiotemporal crowdsourcing resources to UAVs is notable owing to numerous decision variables,which increase linearly with the duration of the service.Furthermore,the relationship between performance metrics and decision variables is intricate,requiring adherence to quality of service requirements.This problem is formalized as a Markov decision process(MDP),providing a structured approach to model the decision-making scenario faced by UAVs in a dynamic environment.To solve this MDP,we employ the soft actor critic(SAC)algorithm,an advanced deep reinforcement learning method known for its sample efficiency and stability.The SAC algorithm is adept at handling the continuous action spaces typical of UAV flight paths and power control problems,making it particularly well-suited for our application.We rigorously test our proposed methods in scenarios involving multiple UAVs,demonstrating the algorithm's effectiveness in managing the spatiotemporal allocation of resources.Our results show that the SAC algorithm achieves faster convergence speed and better solutions than existing state-of-the-art methods,such as the twin delayed deep deterministic policy gradient(TD3)and the deep deterministic policy gradient(DDPG)algorithms.Furthermore,the paper delves into the strategic selection of the optimal number of UAVs to balance the trade-offs between coverage,energy consumption,and operational efficiency.By analytically and empirically examining the impact of the UAV fleet size on system performance,we provide insights into configuring UAV networks to achieve optimal outcomes in terms of AoI,energy management,and security.In conclusion,our research introduces a robust and intelligent framework for UAV resource allocation.The demonstrated efficacy of the SAC algorithm in this context paves the way for its future application in other domains where secure,efficient,and intelligent resource management is paramount.关键词
无人机/工业能源管理/时空众包/资源分配/深度强化学习Key words
unmanned aerial vehicle/industrial energy management/spatiotemporal crowdsourcing/resource allocation/deep reinforcement learning分类
电子信息工程引用本文复制引用
刘娅汐,李旭龙,霍佳皓,皇甫伟..面向工业场景的无人机时空众包资源分配[J].工程科学学报,2025,47(1):91-100,10.基金项目
国家自然科学基金区域基金重点资助项目(U22A2005) （U22A2005）
国家自然科学青年基金资助项目(62301028,62306030) （62301028,62306030）
中国博士后科学基金第73批面上资助项目(2023M730218) （2023M730218）
