Scribd
Upload
78 views5 pages

Kalman Update Equations

1. The equations describe the Kalman filter process which includes prediction and correction steps. 2. It provides an example of using the Kalman filter to estimate position based on initial position, velocity, and time observations. 3. The example calculates the predicted state ahead, error covariance, Kalman gain, updated estimate, and updated error covariance through the Kalman filter process to estimate the next position as 40.0124 km.

Uploaded by

har.san
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
78 views5 pages

Kalman Update Equations

1. The equations describe the Kalman filter process which includes prediction and correction steps. 2. It provides an example of using the Kalman filter to estimate position based on initial position, velocity, and time observations. 3. The example calculates the predicted state ahead, error covariance, Kalman gain, updated estimate, and updated error covariance through the Kalman filter process to estimate the next position as 40.0124 km.

Uploaded by

har.san
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

Project the State ahead:

x^
k1+ Buk
^x
k =A

TIME UPDATE
(PREDICTION)

Project the error Covariance ahead:


T
P
k = AP k1A +Q

Compute the Kalman Gain:


T
P
k H + R
H

P k H T
K k =

Update the estimate via Zk

x^
k
z k H
^x
k +Kk
^x k =

Update the error Covariance

Pk
Pk =( I K kH )

Example:

MEASUREMEN
T UPDATE
(CORRECTION
)

The eqations for kalman filter are


1.

x k = Ax k1+ Buk1 +w k

2.

z k =Hx k + v k

(Measurement)

The following are the Measurements Observed.


Time (ms)
(k)
Position
Velocity
(v)

20
40
80

The equation that we get :

Posk+1 =Posk +vT


Where

x k =Posk

and

H=1

x^
k1+vT
^x
k =

** We try to decrease Error Covariance.

Pk H
H

T
Pk H
K k =

P
k +R

P
k
K k =

x^
k
z k
^x
k +Kk
^x k =

P
k
Pk =( I K k )

H=1 ,

R=.01

Position = 40Km
Velocity = 80 KmpH
Time = 20 sec = (20/3600) Hr

Posk+1 =40+8020 /3600

^x k = 40.444

.. (1)

2
P
k = E[e ]

where

x^ k
e=x k

e=0.444 4

P
k =0.197136
Therefore
... (2)

H=1

Here

and considering

R=.01

T
P
k H + R
H

P k H T
K k =

K k =0.9517 . (3)

x^ k
z k

^x k + K k
^x k =

Where

z k =40(Observed)

^x k =40.444 +0.9517 ( 4040.444 )

^x k =40.0214

.. (4)

P
k
Pk =( I K k )
Pk =( I 0.9517 )0.197136
Pk =0.0095 . (5)
Therefore the (One Step) first Estimates are:
State Ahead
Error Covariance Ahead
Kalman Gain

x^ k

(Prediction)

P
k

(Prediction)

Kk

40.444
0.197136
0.9517

(Update)

Update Estimate

^x k (Update)

40.0214

Error Covariance

Pk (Update)

0.0095

Now the Observation/Position Estimated is:

40.0124 Km

You might also like