| Underwater target tracking is a research hot spot in the field of Marine development and de-fense.Target tracking based on underwater sensor networks have the advantages such as low cost,self-organizing structure and high fault-tolerance,but due to the slow spread velocity of sound waves,propagation delay of signal transmission is large and unnegligible,which leads to asyn-chronous measurements of underwater nodes.The core idea of this paper is to solve the problem of target tracking based on asynchronous measurements of underwater sensor networks.We de-signed several fusion algorithms under different scenarios.The main work and contributions of this dissertation are as follows:(1)For active tracking of scenarios,the active tracking algorithm of underwater target is stud-ied when the measurements are in sequence measurements(ISM).Two fusion algorithms are de-signed:asynchronous fusion method based on time registration and asynchronous fusion method based on state prediction.Simulation result shows that the affection of sampling frequency to the two asynchronous fusion methods.(2)For passive tracking scenarios,the passive tracking algorithm of underwater target based on in sequential measurement(ISM)is studied,a time delay estimation algorithm in passive detec-tion is proposed.Then the asynchronous measurement fusion algorithm is realized.Simulation results show that propagation delay has a greater impact on target tracking accuracy when target speed increase,or the number of nodes increases,it is at that time that considering propagation delay will be more meaningful.(3)To solve the asynchronous fusion problem of out of sequential measurement(OOSM),an OOSM fusion algorithm based on GM-PHD is designed.Negative-time updating of one step lag measurement is carried out under the random finite sets,and the asynchronous fusion framework of OOSM from multi-sensor is given,which can be used to deal with both ISM and OOSM.Sim-ulation results show that the algorithm is reliable. |
PDF Full Text Request