# retrodict

Retrodict filter to previous time step

## Description

The `retrodict`

function performs retrodiction, predicting
the state estimate and covariance backward to the time at which an out-of-sequence measurement
(OOSM) was taken. To use this function, specify the `MaxNumOOSMSteps`

property of the filter as a positive integer. After using this function, use the `retroCorrect`

function to update the current state estimates using the OOSM.

`[`

retrodicts the filter by time `retroState`

,`retroCov`

] = retrodict(`filter`

,`dt`

)`dt`

, and returns the retrodicted state and
state covariance. The function also changes values of the `State`

and
`StateCovariance`

properties of the filter object to
`retroState`

and `retroCov`

, respectively.

`[___,`

also returns the status of the retrodiction `retrodictStatus`

] = retrodict(___)`retrodictStatus`

as
`true`

for success and `false`

for failure. The
retrodiction process can fail if the length of the state history stored in the filter
(specified by the `MaxNumOOSMSteps`

property of the filter) does not
cover the request time specified by the `dt`

input.

## Examples

## Input Arguments

## Output Arguments

## More About

## References

[1] Bar-Shalom, Y., Huimin Chen, and
M. Mallick. “One-Step Solution for the Multistep out-of-Sequence-Measurement Problem in
Tracking.” *IEEE Transactions on Aerospace and Electronic Systems* 40,
no. 1 (January 2004): 27–37. https://doi.org/10.1109/TAES.2004.1292140.

## Extended Capabilities

## See Also

**Introduced in R2021b**