摘要
为了消除机动目标多模型跟踪中人为因素对跟踪精度的影响,提出了一种新的基于时变马尔科夫转移概率的机动目标多模型跟踪算法.该算法通过对Baum辅助函数的最大化实现隐马尔科夫模型状态转移概率的参数估计,并将估计结果用于交互式多模型算法的设计中,构造出时变马尔科夫状态转移概率的交互式多模型算法,有效地降低了人为因素对机动目标跟踪精度的影响.通过对一个机动目标的跟踪对比,说明了该算法比传统的交互式多模型算法具有更小的跟踪误差和良好的模型跟踪概率.
To decrease the man-made effects on target tracking accuracy, a new maneuvering target tracking algorithm is presented, in which the Baum's auxiliary function is maximized to estimate the transition probabilities of hidden Markov model, the interacting multiple model (IMM) tracking algorithm based on time-varying Markov state transition probabilities is designed by means of the above procedure, and the maneuvering target tracking accuracy is increased efficiently. The simulation results for a maneuvering target tracking example indicate that the better tracking model probabilities can be obtained with the proposed algorithm. And in comparison with that of the conventional IMM algorithm, the tracking errors of position and speed are decreased obviously.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2003年第8期824-828,共5页
Journal of Xi'an Jiaotong University
基金
国家重点基础研究发展规划"九七三"资助项目 (2 0 0 1CB3 0 940 4).
关键词
隐马尔科夫模型
转移概率
多模型
跟踪
Computer simulation
Errors
Maneuverability
Markov processes
Monte Carlo methods
Probability distributions
Time varying systems
