基于截断正态概率模型的改进目标跟踪算法

An Improved Target Tracking Algorithm Based on Truncation Gauss Probability Model

在应用NIFPAT算法对机动目标跟踪时存在计算量大和要求网络训练两个弱点。分析了截断正态概率模型和NIFPAT算法后 ,把截断正态概率模型与NIFPAT算法思想相结合 ,提出了基于截断正态概率模型的改进目标跟踪算法。该算法根据目标机动情况自动调整系统方差 ,在附加较小计算量的情况下 ,能取得较好的跟踪效果。仿真实验表明 ,该算法能有效地提高对机动目标的跟踪精度。

In tracking a maneuvering target by NIFPAT algorithm,there are two shortcomings i.e. great computational burden and dependence on network training. After the truncation Gauss probability model and NIFPAT algorithm are analyzed, an improved target tracking algorithm based on the truncation causs probability model is presented by combining the truncation Gauss probability model with the idea of NIFPAT algorithm is this paper. The algorithm can automatically adjust system variation according to target maneuver. With a little computation increasing, a better tracking effect can be obtained. The simulation test shows that the tracking accuracy on a maneuvering target has been improved by using the algorithm and proves that the algorithm is valid.