基于改进EKF的无人机SOC估算分析
|
摘要
多旋翼无人机的工作情况要求采用精度很高的电池剩余电量(SOC)估算方法,而常用的安时法因其特性在该领域中无法做到精确估算,针对多旋翼无人机的工作特性,在戴维南电池模型的基础上,利用扩展卡尔曼滤波法(EKF)估算多旋翼无人机电池剩余电量,并为提高EKF算法的跟踪效果,加入实时调整因子,改进EKF算法。实验结果表明采用扩展卡尔曼滤波算法估算多旋翼无人机的SOC值有较好效果,改进后的算法相比改进前的算法精度又提高了25%,可以有效解决多旋翼无人机电池SOC值无法精准估算的问题。
The operation of a multi-rotor UAV requires a highly accurate battery residual capacity(SOC) estimation method, and the commonly used current integration method cannot be accurately estimated in this field due to its characteristics. In order to find a method for accurately estimating the SOC value of a multi-rotor UAV battery, based on the full analysis of the working environment of the multi-rotor UAV, the extended Kalman filter method was applied to estimate the SOC value of the UAV battery, and here based on the introduction of dynamic change gain,the tracking effect estimated by the extended Kalman filter method was improved. The experimental results show that the extended Kalman filter algorithm is effective in estimating the SOC value of the multi-rotor UAV. The improved algorithm improves the accuracy of the algorithm before the improvement by 25%, which can effectively solve the SOC value of the multi-rotor UAV battery and the problems that cannot be accurately estimated.
The operation of a multi-rotor UAV requires a highly accurate battery residual capacity(SOC) estimation method, and the commonly used current integration method cannot be accurately estimated in this field due to its characteristics. In order to find a method for accurately estimating the SOC value of a multi-rotor UAV battery, based on the full analysis of the working environment of the multi-rotor UAV, the extended Kalman filter method was applied to estimate the SOC value of the UAV battery, and here based on the introduction of dynamic change gain,the tracking effect estimated by the extended Kalman filter method was improved. The experimental results show that the extended Kalman filter algorithm is effective in estimating the SOC value of the multi-rotor UAV. The improved algorithm improves the accuracy of the algorithm before the improvement by 25%, which can effectively solve the SOC value of the multi-rotor UAV battery and the problems that cannot be accurately estimated.
引文
[1]乐淑玲,王云,刘诗鸾,等.新型混合动力无人飞行器气动特性数值分析[J].计算机仿真,2014,31(6):87-91.
[2]李鹏,招启军.悬停状态倾转旋翼/机翼干扰流场及气动力的CFD计算[J].航空学报,2014,35(2):361-371.
[3]李硕,缑林峰.飞行器动力系统多变量控制结构设计研究[J].测控技术,2015,34(2):88-90.
[4]李百华,郭灿彬,钟其水,等.电动汽车锂电池戴维南等效电路模型参数辨识研究[J].微型机与应用,2017,36(1):83-85.
[5] PAZ L M,TARDS J D,NEIRA J.Divide and conquer:EKF SLAMin O(n)[J]. IEEE Transactions on Robotics,2008,24(5):1107-1120.
[6] LI M Y,MOURIKIS,ANASTASIOS I. High-precision,consistentEKF-based visual–inertial odometry[J]. International Journal ofRobotics Research,2013,32(6):690-711.
[2]李鹏,招启军.悬停状态倾转旋翼/机翼干扰流场及气动力的CFD计算[J].航空学报,2014,35(2):361-371.
[3]李硕,缑林峰.飞行器动力系统多变量控制结构设计研究[J].测控技术,2015,34(2):88-90.
[4]李百华,郭灿彬,钟其水,等.电动汽车锂电池戴维南等效电路模型参数辨识研究[J].微型机与应用,2017,36(1):83-85.
[5] PAZ L M,TARDS J D,NEIRA J.Divide and conquer:EKF SLAMin O(n)[J]. IEEE Transactions on Robotics,2008,24(5):1107-1120.
[6] LI M Y,MOURIKIS,ANASTASIOS I. High-precision,consistentEKF-based visual–inertial odometry[J]. International Journal ofRobotics Research,2013,32(6):690-711.