RcppTests

Doing the essential tests involving Kalman filter implementations

Easy Test

Open and run Easy.R

Medium Test

Open and run Medium.R. The data pos.txt is used in Medium.R

Hard Test

Install package myKalman_1.0.tar.gz. Example to test the package:

# An example, the same as used in the Kalman filtering example in the RcppArmadillo package
# See http://dirk.eddelbuettel.com/papers/RcppArmadillo-intro.pdf
  
  library(myKalman)
  data(pos)
  z <- pos
  dt <- 1
  F <- matrix( c( 1, 0, dt, 0, 0, 0,  # x
                0, 1, 0, dt, 0, 0,   # y
                0, 0, 1, 0, dt, 0,   # Vx
                0, 0, 0, 1, 0, dt,   # Vy
                0, 0, 0, 0, 1,  0,   # Ax
                0, 0, 0, 0, 0,  1),  # Ay
                6, 6, byrow=TRUE)
  B <- matrix(0, 6, 6) # not relevant for this example
  N <- nrow(pos)
  Q <- diag(6)
  H <- matrix( c(1, 0, 0, 0, 0, 0,
               0, 1, 0, 0, 0, 0),
               2, 6, byrow=TRUE)
  R <- 1000 * diag(2)
  u <- matrix(0, N, 6) #not relevant for this example
  xinit <- matrix(0, 6, 1)
  Pinit <- matrix(0, 6, 6)

  results <- Kalman(F,B,Q,H,R,z,u,xinit,Pinit)

  # Plot of the original and estimated trajectories
  plot(z[,1],z[,2],pch=1,col=1,xlab="x",ylab="y")
  points(results$Y[,1],results$Y[,2],pch=2,col=2)
  legend("topleft",legend=c('Original','Estimated'),col=1:2,pch=1:2)