Group target Poisson multi-Bernoulli hybrid filtering method, system and filter

The invention relates to a group target Poisson multi-Bernoulli hybrid filtering method and system and a filter, and the method carries out the deep improvement design of the recursion processing process of a prediction step and an updating step on the basis of the recursion process of a standard PMBM filter, thereby changing the previous filtering processing flow, and compared with the traditional technology, the method provided by the invention has the advantages that the efficiency is improved, and the cost is reduced. A measurement driving processing mode is adopted, the new target state is no longer assumed to be accurately known in advance, and the new target is initialized by using the current frame measurement, so that the recursive processing process of the whole filter does not need to depend on the accurate filtering initial state any more, and the filtering accuracy is improved. Therefore, efficient Poisson multi-Bernoulli hybrid filtering of the group target can be realized, and the applicability