论文部分内容阅读
作者描述了一种使用模糊逻辑理论推广的交互式多模型(IMM)算法。每一个卡尔曼滤波器都被认为是局部有效的,适用于滤波调节值所确定的目标加速度空间的区域。每个卡尔曼滤波器的有效性是由模糊集合和目标的加速度估值所决定的。用选择适当的模糊集合的叠区间判断接近加速度估值要求的模型的子集。仿真结果包括了典型的机动目标情况,它说明了不同形式的模糊集合的作用。这些证明了这种方法在估计精度和计算量两方面都比IMM优越。
The authors describe an interactive multi-model (IMM) algorithm that is generalized using fuzzy logic theory. Each Kalman filter is considered locally valid and applies to the area of the target acceleration space as determined by the filter adjustment value. The effectiveness of each Kalman filter is determined by the fuzzy set and the target’s acceleration estimate. A subset of the models required for near-acceleration estimation are judged by the overlap between selected fuzzy sets. The simulation results include the typical case of maneuvering targets, which illustrates the effect of different forms of fuzzy sets. These prove that this method is superior to IMM in both estimation accuracy and computational load.