一类新型机动目标跟踪算法

阐述了当跟踪非机动目标时,传统的Kalman滤波可以得到很好的跟踪精度。但是当日标机动时,传统的Kalman滤波不能对目标的突然变化做出及时的改正和预测,因此跟踪精度很差,甚至出现丢失目标的情况。文中采用的基于截断正态概率模型的改进自适应目标跟踪算法, 其结构和计算简单,鲁棒性好,较好地解决了使用Kalman滤波带来的不足。

A Novel Algorithm for Maneuvering Target Tracking

While tracking non-maneuvering targets, the conventional algorithms based on Kalman filter are competent. However, they are no longer valid in cases of occurrence of target maneuvering. They could not response quickly to the sudden change of targets. Hence, neither can they adjust system models with short latency, nor make correct prediction. In applications for tracking maneuvering targets, they cannot provide sufficient rigorous, and always miss the target while tracking. This paper proposes a novel-tracking algorithm based on adaptive tracker. It adopts the truncate norm distribution model to improve the performance of the adaptive tracker. The structure and computation process are quite simple. Experiments show that it gives good robustness to delude the disadvantages of using Kalman filter directly in cases of tracking maneuvering targets.