论文部分内容阅读
3 贝叶斯滤波 我们现在回到状态估计问题上。接着上节中描述的分离原则,我们认为观测已被互联,也就是说,划分成子集,每个子集对应于一个目标。状态估计问题要采用已知集合中的观测来估计目标状态。我们希望在这一点上有所区别。我们把目标叫作我们正在估计其状态的实际对象。例如,在水面舰艇跟踪问题中,目标是舰艇。另一方面,在用互联算法进行分离的情况下,航迹是被联结为单个子集的观测集。区别是:目标是真实对象,航迹是由数据融合系统建立的结构。
3 Bayesian filtering We now return to the state estimation problem. Following the separation principle described in the previous section, we consider observations to be interconnected, that is, divided into subsets, each of which corresponds to a target. State Estimation Problems Use the observations in known sets to estimate the state of the target. We hope to make a difference at this point. We call the goal a real object for which we are estimating its status. For example, in the case of surface ship tracking, the target is a ship. On the other hand, in the case of the separation using the interconnection algorithm, the track is an observation set linked into a single subset. The difference is: the target is the real object, and the track is the structure established by the data fusion system.