This might not be the answer to your question, but If you are looking for beginner-friendly examples of the Kalman filter, I can always recommend "Probabilistic Robotics" (see http://www.probabilistic-robotics.org/).

They explain Bayesian in general and Kalman filters in particular using easy to understand use cases from robotics.

Dear Desmond Djoemena,

I suggest you to see links and attached file on topic.

http://lalas.github.io/quantitativeThoughts/r/2014/09/01/dlmTutorial.html

https://kevinkotze.github.io/ts-4-tut/

http://lenkiefer.com/2018/06/10/kalman-filter-for-a-dynamic-linear-model-in-r/

https://towardsdatascience.com/state-space-model-and-kalman-filter-for-time-series-prediction-basic-structural-dynamic-linear-2421d7b49fa6

Article Kalman Filtering in R

Best regards

The kalman filter needs a mathematical model as a represenatation of the problem.

yt = Ft * xt + vt

xt = Gt*xt-1 + wt

The prior distribution @ t=0 is modeled as a Gaussian distribution with mean m0 and variance C0: N(m0,C0)

vt: process noise, modeled as Gaussian dist. N(0, Vt)

wt: measurement noise, modeled as Gaussian dist. N(0, Wt)

As an example you can see target tracking scenerios:

http://ais.informatik.uni-freiburg.de/teaching/ws11/robotics2/pdfs/rob2-19-tracking.pdf

For constant velocity model:

state: x = [x, y, vx, vy] -> positions x and y, velocities x and y

Ft is a matrix that holds mathematical equations, x1 = x0 + vx*t, y1 = y0 + vy*t

Gt is a matrix, sometimes denotes as H as in the tutorial, for relating state to observations

Vt and Wt are constants for noise matrices Q and R, e.g. Q = Vt * I(4)

m0 and C0 are just initial values.

m0 is the initial state consists of initial positions and velocities

C0 is a constant for initial covariance matrix P0