Research On Target Tracking Problem For Mobile Sensor Network Based On Flocking Control

With the rapid development of wireless communication technology and electronic equipment,sensor networks have been widely used as an important monitoring platform in target tracking.Due to the limited detection range of the sensors,it is difficult for the system to effectively track and monitor remote targets.Therefore,in the research of mobile sensor network,coupled distributed estimation and control problem is discussed.Through controlling the network to follow the target motion,real-time monitoring of the target is ensured.In this paper,the target tracking problem for mobile sensor network based on flocking control is studied.The existing algorithm is improved for velocity-varying target.Then,the velocity-varying target tracking algorithm is extended by network structure expansion,avoiding obstacles and considering the situation where the target is constrained.The contents can be divided into four parts:1.The velocity-varying target tracking algorithm for mobile sensor network based on flocking control is researched.The state estimates of Kalman Consensus Filter are used as the guidance feedback term in the control input.The theoretical analysis and simulation results show that all estimators asymptotically reach a consensus on the state estimates,the position and velocity of the center of the sensor network converge to those of the target,and that all sensors move with the target speed.2.The network connectivity and detection function of all sensors are needed for the velocity-varying target tracking algorithm to be effective.By introducing the edge-added lag rule and improving the artificial potential energy function in the target tracking algorithm,the connectivity of the sensor network is ensured.The algorithm is proved to be feasible through theoretical analysis.The artificial potential energy function terms with low complexity is designed to improve the computational efficiency of the algorithm.On the basis of preserving network connectivity,the situation while only a fraction of sensors could obtain target information is discussed.Under such situation,the preconditions for the tracking target system to be stable is given.The simulation results verify the feasibility of the algorithm and analyze how the ratio of sensors with target information will affect the estimation accuracy and control effect.3.Based on the projection method,null space method and active set method,the state estimation filtering algorithm with constraints is studied.The target tracking algorithm for mobile sensor network with constraints is proposed through using the constrained estimates as the guidance feedback term in the control input.The simulation results show that by utilizing the prior constraint information,the tracking precision of the network is increased and the control effect is optimized.4.The target tracking algorithm for mobile sensor network with obstacle avoidance is studied.By creating virtual agents on all nearby obstacles,the network avoids obstacles while tracking the target.By adjusting the warning distance,the control input is reduced.The simulation results verify the feasibility of the algorithm.