基于模糊逻辑的改进自适应IMM跟踪算法

The interacting multiple model algorithm (IMM) and fuzzy systems-based interacting multiple model algorithm (FIMM) are practical maneuvering target tracking algorithms with fixed model sets. When a large number of models are needed to cover all possible maneuver cases, it will lead to a surge in computation, and may even lead to unnecessary model competition, thus reducing the tracking performance. In view of this defect, an improved fuzzy-logic adaptive IMM algorithm (FAIMM) is proposed, which adopts a nonlinear mapping method of model probabilities to screen the subset of models in real time, eliminate useless models, and increase the weight of useful models. Besides, it can adjust the process noise level automatically through a fuzzy system, so that the algorithm has stronger adaptive ability to different target maneuvering modes. Simulation results show that the tracking performance of the proposed algorithm is better than that of IMM and FIMM, and can match the target maneuvering mode better.