Research on dynamic target tracking algorithm in mobile sensor networks

Abstract

With the increasing complexity of networked cooperative detection environment, new targets appear and leave in the detection area, and the mobility of targets is increasing. The emergence of sensors makes the sensors in the sensor network have certain motion characteristics. These uncertainties increase the complexity of the problem. At the same time, the dynamic selection of sensor resources needs to consider the needs of different priority tasks, platform survival risk, sensor cost and other factors. In view of these problems to be solved, based on the qualitative synthesis theory of predictive control and stochastic dynamic programming optimization theory, the dynamic sensor selection decision theory and method based on multi-step forward decision are explored. This paper studies the dynamic sensor strategy based on multi-step forward decision scheduling, and carries out the research on the optimization decomposition method of sensor selection under different structural characteristics, the dynamic optimization method of sensor selection under complex and uncertain environment, and the multi-objective optimization method of dynamic sensor selection under heterogeneous targets, so as to explore a new way of dynamic sensor selection algorithm design and performance analysis under networked collaborative tracking A new way. The research of dynamic target tracking algorithm in mobile sensor networks has important theoretical and engineering significance to improve the resource efficient utilization and the overall target tracking performance of sensor networks.