基于随机集理论的多个声目标融合跟踪

针对杂波环境下，采用多个被动声传感器跟踪多个声目标的应用场合，建立了多个声目标跟踪的随机有限集模型，采用概率假设密度(probability hypothesis density, PHD)粒子滤波对该模型进行求解。针对PHD滤波器只适用于单传感器的问题，提出了一种实现多个声传感器融合跟踪的方法。该方法在序贯PHD滤波器的基础上进行改进，提高了目标检测率，通过仿真实验验证了该方法的有效性。

Multi-acoustic-target fusion tracking based on random set theory

To investigate the problem of tracking multiple acoustic targets with multiple passive acoustic sensors in a clutter environment, the multi-target tracking model is built based on the random finite set. The probability hypothesis density (PHD) particle filter is used to solve the model. Aiming at the problem of the PHD filter is mainly for the single sensor case only, a method of multi-sensor fusion tracking is proposed, which takes advantage of the sequential PHD filter and could increase the detection rate. Finally, an example of data simulation demonstrates the effectiveness of the approach.