基于粒子滤波的机载目标跟踪系统设计

Object tracking system for MUAV based on particle filter

基于微小型机载成像跟踪系统设计思想及需求,设计并实现了以高性能的DSP芯片TMS320-DM642为核心处理器,结合可编程逻辑器件CPLD和FPGA的实时图像跟踪处理平台。平台采用基于粒子滤波的目标跟踪算法,实现对目标的实时跟踪。采用卡尔曼滤波器,提高了粒子的利用效率,在改进了算法实时性的同时解决了图像跟踪系统的延时性问题,提高了跟踪系统的稳定性。算法仿真结果表明,与传统相关匹配算法相比,基于粒子滤波的跟踪算法具有更好的鲁棒性和实时性,能满足机载成像跟踪系统实时图像跟踪的要求。

To meet the requirements of micro airborne electro-optical tracking system for MUAV,a real-time image tracking processing system is introduced.The system is constructed by the high performance DSP chipset TMS320 DM642 as the core digital processor,integrated with the reprogrammable logic chipset CPLD and FPGA as the image preprocessing chipset.Particle filter is implemented as object tracking algorithm.Kalman filter is applied to increasing the speed of computation and solving the delay of image tracking system.Finally,the experiment is given to demonstrate that the algorithm work well in real-time and robust.