Multi-Sensor Cooperative Tracking Using Distributed Nash Q-Learning

摘要：

　　Traditional target tracking algorithm has a disadvantage of excessive dependence on the environment model.Thus a multi-sensor cooperative tracking method using distributed Nash Q-learning was proposed.Distributed Nash Q-leaming with model-free was firstly described.Then sensor action and reward function were defined,which both are very crucial to the learning.Sensor action was only subjected to angle control,and reward function was given by calculating the trace of one time-step prediction error covariance.Nash tragedy can not be directly calculated,therefore,a probability statistics method using Bayesian inference was used to update the Q function.Simulation of passive tracking merely with angle measurements shows that this algorithm can enhance the adaptation to environment change and the tracking accuracy.

关键词： Reinforcement learning Nash Q-learning Target tracking Extended Kalman filtering Multi-sensor cooperation Distribution

作者: Jia Cai Changqiang Huang Haifeng Guo

作者单位: Aeronautics and Astronautics Engineering Institute,Air Force Engineering University,Xian,China

会议类型: 国际会议

会议名称: the 2012 International Conference on Manufacturing Engineering and Automation (2012年制造工程与自动化国际会议(ICMEA2012))

会议地点: 广州

会议语种:英文

页码: 1475-1478

在线出版日期: 2012-11-16（万方平台首次上网日期，不代表论文的发表时间）