自动化学报2024，Vol.50Issue(7)：1417-1431,15.DOI:10.16383/j.aas.c230591
扩展目标跟踪中基于深度强化学习的传感器管理方法
Sensor Management Method Based on Deep Reinforcement Learning in Extended Target Tracking
摘要
Abstract
To solve the problem of sensor management in the optimization of extended target tracking(ETT),this paper proposes a sensor management method based on deep reinforcement learning(DRL)by modeling the exten-ded target based on random matrices model(RMM).First,in the theoretical framework of partially observed Markov decision process(POMDP),a elementary method of sensor management for extended target tracking based on twin delayed deep deterministic policy gradient(TD3)algorithm is presented.After that,the Gaussian Wasser-stein distance(GWD)is used to calculate the information gain between the prior probability density and the pos-terior probability density of the extended target,which is used to comprehensively evaluate the multi-feature estim-ation information of the extended target,and then the information gain is used as the reward function of TD3 al-gorithm.Furthermore,the optimal sensor management scheme based on deep reinforcement learning is decided by the derived reward function.Finally,the effectiveness of the proposed algorithm is verified by constructing an ex-tended target tracking optimization simulation experiment.关键词
传感器管理/扩展目标跟踪/深度强化学习/双延迟深度确定性策略梯度/信息增益Key words
Sensor management/extended target tracking(ETT)/deep reinforcement learning(DRL)/twin delayed deep deterministic policy gradient(TD3)/information gain引用本文复制引用
张虹芸,陈辉,张文旭..扩展目标跟踪中基于深度强化学习的传感器管理方法[J].自动化学报,2024,50(7):1417-1431,15.基金项目
国家自然科学基金(62163023,62366031,62363023,61873116),甘肃省教育厅产业支撑计划项目(2021CYZC-02),2024年度甘肃省重点人才项目资助Supported by National Natural Science Foundation of China(62163023,62366031,62363023,61873116),Gansu Province Edu-cation Department Industrial Support Project(2021CYZC-02),and Key Talent Project of Gansu Province in 2024 （62163023,62366031,62363023,61873116）
