模型转移概率自适应的交互式多模型跟踪算法

Interacting Multiple Model Algorithm with Adaptive Markov Transition Probabilities

该文利用量测中所包含的当前模式信息,实现了马尔可夫转移概率的实时估计,并将估计结果用于交互式多模型跟踪算法(IMM)的设计中,构造出参数自适应的交互多模型跟踪算法(PAIMM),有效降低了人为因素的影响。通过一个跟踪机动目标的仿真实例,说明PAIMM算法的有效性。

A estimator of the time-varying Markov state transition probabilities is presented , which is based on the measurements. Then the Parameter Adaptive Interacting Multiple Model （PAIMM） is designed by adopting the above estimator. In comparison with that of the conventional IMM algorithm, the tracking performance of PAIMM is better in the simulation of tracking a maneuvering target.