Path: Common/Estimation
% Rauch-Tung-Striebel smoothing This assumes a discrete model of the form: x[k] = a[k-1]x[k-1] + b[k-1]u[k-1] y[k] = h[k]x[k] You pass all of the filter matrices that were already generated by a Kalman Filter. Type RTSS for a demo of a double integrator system. -------------------------------------------------------------------------- Form: [mS, pS] = RTSS( mP, pP, m, p, a ) -------------------------------------------------------------------------- ------ Inputs ------ mP (n,:) Mean vector from KF prediction step pP (n,n,:) Covariance matrices from KF prediction step m (n,:) Mean vector from KF p (n,n,:) Covariance matrices from KF a (n,n,:) State transition matrices ------- Outputs ------- mS (n,1) Smoothed mean vector pS (n,n) Smoothed covariance matrix --------------------------------------------------------------------------
Common: CommonData/SwooshWatermark Common: Estimation/KFPredict Common: Estimation/KFUpdate Common: General/CellToMat Common: General/MatToCell Common: General/Watermark Common: Graphics/NewFig Common: Graphics/Plot2D Common: Graphics/PltStyle Common: Graphics/TimeLabl
Back to the Common Module page