A Decision-Based Model and Algorithm for Maneuvering Target Tracking

摘要

The decision-based models for maneuvering target tracking were studied in this paper. Focusing on the problem of dissatisfied with single model tracking and the optimal model-set is difficult to design of the multiple-model (MM) algorithm, we modified the "current" model, and proposed an adaptive single model (ASM) to track angular motion. The united velocity-acceleration estimation-based and direct measurement-based decision algorithms for model-switching were provided and discussed. At the same time, we discussed the question of the thresholds setup, and gave a significance test method based on hypothesis testing theory. Simulation results show that the tracking performance of this new adaptive single model is stationary, and is improved much more than the "current" model, also better than the interacting multiple-model (IMM) for strong maneuvering target tracking. Furthermore the computational load of the new model can be less than that of IMM.