Research On Composite Tracking Of Multisensor And Multitarget

Nowadays, with the development of technology, the circumstance of modern war has changed a lot. The traditional flat warfare is difficult to meet the actual combat's need, so it is necessary to turn the research focus to the Network Centric Warfare (NCW) in order to tread on the heels of the development trend of the future warfare. The biggest characteristic of the NCW is to actualize the Cooperative Engagement Capability (CEC) by the highly real time share of the warfare development situation. The Composite Tracking Algorithm research in this paper is one of the three important applications in the tactics of CEC. The directing principle of this subject is, in according with the practical project, to research the Composite Tracking Algorithm in the blind area and common area.Firstly, the maneuver targets'movement model and the maneuver targets tracking algorithm are studied. Simulation results demonstrate that the IMM and Current Statistical Algorithm can deal with the maneuver targets tracking very well. The concept and the connotation of the Composite Tracking Algorithm are put forward according to the desire of the CEC. The fusion tracking algorithm in the common area of multi-sensor network, the extending track algorithm and the relay tracing algorithm in the blind area are expanded grounded on the frame of the composite tracking.The fusion tracking algorithm under common area, Fusion with non-equal dimension, asynchronous multi-sensor track fusion and the data fusion based on the combination of radar and infrared imaging sensor are studied in this paper. In factual data fusion system of radar fence, each radar'sampling time is not synchronous and the sampling period is also not identical at majority instances, also in the structure of distributed multi-sensor data fusion, each sensor may have different observation dimension; that is to say, each sensor may make use of different model to observe the same target. So these researches in the paper make sense in practice. In consideration of the asynchronous problem of data from radar and infrared sensor, a time aligning based on last–square technique and fusion method are investigated in this paper. The fusion tracking accuracy is better than the single sensor tracking accuracy. An extended kalman filtering with interacting multiple model maneuvering target tracking algorithm is presented considering the feature of this combined guiding system when targets maneuvering. This algorithm combines the advantages of the extended kalman filtering and the interacting multiple model algorithms. Simulation results indicated that the whole performance of the algorithm presented in this paper is better than the extended kalman filtering algorithm.Finally, extending track algorithm and the relay tracking algorithm are studied in this paper. Two ways of extending track are presented: one is linear movement model extending track based on maneuver detecting, and the other is turning movement model extending track based on estimation of parameter of model. The extended track is then modified by the local systematic errors, and then relay tracking algorithm is worked to produce composite track. It is shown by the theory and calculation that extended tracking, based on the adaptive fittings of the systematic errors and covariance matrices, can in some degree resist the influences of the systematic errors on the estimated states of relay tracking.The research of Composite Tracking Algorithm remains the primary phase, only the notation is researched, an integrated system does not form at the moment. Based on Network Centric Warfare (NCW), a new concept of composite tracking in the complex environments is brought forward and the Composite Tracking Algorithm is researched. In this paper we present a creative method based on above research for blind area tracking, computer simulation show the effect of algorithm.