kalmants.m

KALMAN FILTERING FOR FUZZY DYNAMIC SYSTEMS

function [EstAve, MeasAve] = KalmanTS(sigmaX, sigmaY, fOptimal)

% function [EstAve, MeasAve] = KalmanTS(sigmaX, sigmaY, fOptimal)
%
% Monte-Carlo simulation of the truck-trailer system.
% INPUTS
%   sigmaX = 3-element vector containing the standard deviation of 
%            the state process noise (rad, rad, velocity)
%            三元素向量包括状态过程噪声的标准差
%   sigmaY = 3-element vector containing the standard deviation of
%            the measurement noise (rad, rad, velocity)
%            三元素向量包括测量噪声的标准差
%   fOptimal = flag indicating if we should use the optimal uncoupled
%              Kalman filter.  If 1, then we use the optimal 
%              Kalman filter.  If 0, then we use the suboptimal decoupled
%              Kalman filter.
% OUTPUTS
%   EstAve = average RMS state estimation error  %均方根状态估计误差
%   MeasAve = average RMS measurement error      %均方根测量误差
   sigmaX=[1;2;3];sigmaY=[1;2;3];fOptimal=0;             %05-09-27乐加
EstAve = [0 0 0];
MeasAve = [0 0 0];
for i = 1 : 5
   [EstRMS, MeasRMS] = TruckTrailer(sigmaX, sigmaY, fOptimal, 0);
   EstAve = ((i - 1) * EstAve + EstRMS) / i;
   MeasAve = ((i - 1) * MeasAve + MeasRMS) / i;
end
disp(['Average Estimation Error = ', num2str(EstAve)]);
disp(['Average Measurement Error = ', num2str(MeasAve)]);