Group target tracking method based on robust MS-MeMBer filtering

The invention discloses a group target tracking method based on robust MS-MeMBer filtering, and belongs to the technical field of information processing, and the method comprises the steps: initializing parameters, then carrying out the prediction step of robust MS-MeMBer filtering, and generating predicted multi-target density; obtaining a multi-sensor measurement set, and carrying out measurement partition on the multi-sensor measurement set based on an S-D matching algorithm to generate a measurement partition set; calculating the predicted multi-target density and the measurement to-be-divided set to obtain updated multi-target density; obtaining a group structure of the group target through a graph theory algorithm and an adjacent matrix, and obtaining a motion state of the group target through a stochastic differential equation model; and transmitting the group structure and the motion state to a prediction step of robust MS-MeMBer filtering at the next moment for iterative loop, and generating a tracke