Kalman Missing Observation Handling

Problem

Handle a missing observation in a one-dimensional Kalman filter: the state mean is unchanged and the variance grows by the process noise, so return [prior, p + q].

No measurement update happens.

Hint

Skipping an update only inflates uncertainty.

Test cases

case 1

in: [10,1,0.5]

expected: [10,1.5]

case 2

in: [0,2,2]

expected: [0,4]

case 3

in: [5,0,1]

expected: [5,1]

case 4

in: [3,1,0]

expected: [3,1]

Starter codePython

def kalman_missing_observation(prior, p, q):
    # Your code here
    pass

4 test cases · run your code against them in the editor