Kalman Filter

1
Kalman Filter
2
The noisy process model
Internal state noise
Measurement noise
3
(No Transcript)
4
We make an a priori estimate of the state based
on our knowledge about the system upto time tk
Before taking the measurement
5
As soon as we get the measurement, we refine the
a priori estimate to make the a posteriori
estimate
6
A posteriori estimate
7
(No Transcript)
8
Now we should find the optimum K that minimizes
the error covariance
9
For optimality
Kalman gain
10
(No Transcript)
11
Projecting ahead
The best estimate of the next state