基于FPGA目标跟踪检测的二维角度估计方法  

A Method of Two-dimensional Angle Estimation for Target Tracking and Detection Based on FPGA

作　　者：兰初军 宋志勇[1] 张永强 徐勇 刘宇 LAN Chujun;SONG Zhiyong;ZHANG Yongqiang;XU Yong;LIU Yu(ATR Laboratory,National University of Defense Technology,Changsha Hunan 410073,China)

机构地区：[1]国防科技大学ATR国防重点实验室,湖南长沙410073

出　　处：《现代雷达》2024年第11期1-7,共7页Modern Radar

基　　金：国家创新研究群体科学基金资助项目(61921001)。

摘　　要：针对雷达导引头在低信噪比下角度估计精度不高、收敛时间过长等问题,提出一种基于现场可编程门阵列(FPGA)目标跟踪检测的二维角度估计算法。该算法采用和差差多通道联合滤波,在检测的同时将目标的角度信息集成于伯努利滤波器的状态向量进行递归估计。为了缩短角度估计处理时间,采用FPGA流水化设计,并对影响角度估计精度的计算节点,采用“浮点数+定点数”和拆分指数分别计算,以提高角度估计精度。通过仿真与测试实验,与检测后跟踪的传统算法相比,所提算法取得了较好的角度估计性能;此外,该算法单帧处理时间达到了微秒级,角度估计收敛时间短,极大地提高了目标跟踪检测的实时性。Aiming at the issues of low angle estimation accuracy and long convergence time of radar seekers under low signal-to-noise ratio,a two-dimensional angle estimation algorithm for target tracking and detection based on field programmable gate array(FPGA)is proposed in this paper.The sum difference multi-channel joint filtering is adopted by this algorithm,and the angle information of the target is integrated into the state vector of the Bernoulli filter for recursive estimation while detecting.In order to shorten the processing time of angle estimation,the design of FPGA pipelining is adopted,and the calculation of“floating point+fixed point”and exponential classification are used for the calculation nodes that affect the accuracy of angle estimation to improve the accuracy of angle estimation.Through simulation and testing experiments,compared with traditional algorithms for tracking after detection,the proposed algorithm achieves better angle estimation performance.In addition,the processing time of a single frame of the proposed algorithm reaches the microsecond level,and the convergence time of angle estimation is short,which greatly improves the real-time performance of target tracking detection.

关 键 词：跟踪检测 粒子滤波 现场可编程门阵列 伯努利滤波器 角度估计 

分 类 号：TN957.51[电子电信—信号与信息处理]