双传感器概率假设密度滤波解析实现方法

Analytic Implementation Method of the Double-sensor PHD Filter

针对双传感器概率假设密度(PHD)理论的解析实现进行研究。Mahler给出的双传感器PHD理论,由于其中含有抽象的多目标积分,并且其中的二元分割过程计算量十分巨大,所以无法计算机实现。文中在线性高斯混合的假设条件下给出了严格双传感器PHD滤波的递推解析公式,并且通过提出"有效二元分割"算法极大降低了严格理论意义下的双传感器PHD算法的计算复杂度,从而解决了双传感器PHD滤波的计算实现问题。计算机模拟仿真验证了所提出算法的有效性。

The analytic suboptimum solution of the multi-sensor probability hypothesis density （MS-PHD） filter is studied.For the theoretically rigorous MS-PHD filter due to R.Mahler,because it contains the abstract multiple target integral,and the binary segmentation process calculation burden is very heavy,the filter is impossible to implement on computer.Under the linear Gaussian assumptions,an analytic suboptimum solution is given for the theoretically rigorous MS-PHD filter.Furthermore,a method,named ＂Effective Binary Partition （EBP）＂,is proposed for limiting the number of considered partitions to reduce the computational complexity.The EBP method makes it possible to implement the exact MS-PHD filter formulas on computer.Finally,the validity of the proposed algorithm is demonstrated by numerical simulations.