基于FPGA的扩展卡尔曼滤波器设计与仿真  

Design and Simulation of Extended Kalman Filter Based on FPGA

作　　者：刘一康 刘宁[1] 苏中[1] 范军芳[1] LIU Yi-kang;LIU Ning;SU Zhong;FAN Jun-fang(Beijing Key Laboratory of High Dynamic Navigation Technology,Beijing Information Science and Technology University,Beijing 100192,China)

机构地区：[1]北京信息科技大学高动态导航技术北京市重点实验室,北京100192

出　　处：《计算机仿真》2023年第11期236-242,共7页Computer Simulation

基　　金：国家自然科学基金(61801032);北京市自然科学基金(4212003)。

摘　　要：为了进一步提高工程应用中扩展卡尔曼滤波器的实时性及通用性,提出了基于具有高度并行性特点的现场可编程逻辑阵列(FPGA)实现扩展卡尔曼滤波器的方案。依据其并行运算的特点,设计了扩展卡尔曼滤波并行运算流程,有效缩短了运算时间。针对矩阵乘法、状态机设计等难点提出优化实施方案,并采用模块化的设计方法进行了系统软件架构设计,使其更加易于进行测试与升级。最后通过与数值仿真结果进行对比分析,验证了本设计具备良好的运算精度,且有效提高了系统实时性。In order to improve the real-time performance and universality of the extended Kalman filter in engi-neering applications,a scheme to implement the extended Kalman filter based on field programmable logic array(FP.GA)with high parallelism is proposed.According to the characteristics of parallel operation,the parallel operation flow of the extended Kalman filter is designed,which can effectively shorten operation time.In view of the matrix multipli-cation,state machine design and other difficulties,the optimization implementation scheme is proposed,and the modu-lar design method is adopted to design the system software architecture,which makes it easier to test and upgrade.Fi-nally,the comparison with numerical simulation results verifies that the design has good calculation accuracy,which effectively improves the system real-time performance.

关 键 词：扩展卡尔曼滤波器 矩阵运算 状态机 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]