What is the point in a Kalman filter ? Why don't people just come up with a decent estimator in the first place and cut out all the bullshit ?
Lack of computational power or lack of information. In both cases you can only approximate. Look at this. http://en.wikipedia.org/wiki/Kalman_filter
I am aware of wikipedia.
I refer you to the original question. You may assume infinite computational power.
Ok, then the second premise where not all the information is not there, or not to a level of accuracy. A simple example is the Radar, there is always interference with the signal. The filter approximates to find and enemy location, in a similar fashion to the Bayesian stats used in GPS.
i bet it is a very long time before you get laid.
I doubt it McJ, I can get laid when I want to, I have pretty high levels of personal charisma. Bear in mind maths is my specialism really, so I do know the stuff...
I doubt that either personal charisma or maths is your speciality.
Do you see why ?
Maths, I have 2 alevels in it both A's hopefully and am off to take it to degree level probably along with other things. Also I do read quite heavily around the subject.
As for personal charisma, you would be surprised what I have talked to people into over the years, and people do admire and respect me (good ears help in finding out about that one).
The estimations are a process carried out by an approximation algortihm. The filters are needed to clear it up before the algorithm is applied.
Well, that's great news.
I doubt you will do this in a mathematics degree though ( unless you are unlucky, or a big enough idiot to actually choose to).
Anyway, you don't need a 'filter' to 'clear it up' if you have a decent approximation in the first place given the parameters. If the parameters are not accurate due to incomplete information, then all your 'filter' is doing is gaining more information. Which you could just measure anyway.
The state variables are just measurements to calculate parameters for the predictor.
If state variables are accurate then the Kalman filter converges with respect to how good your initial estimator is.
Hence, my question.
Why not put the effort into getting a good predictor, instead of all the bullshit ?