Main Content

Gaussian mixture (GM) PHD filter

The `gmphd`

object is a filter that implements the
probability hypothesis density (PHD) using a mixture of Gaussian components. The filter
assumes the target states are Gaussian and represents these states using a mixture of Gaussian
components. You can use a `gmphd`

filter to track extended objects or
point targets. In tracking, a point object returns at most one detention per sensor scan, and
an extended object can return multiple detections per sensor scan.

You can directly create a `gmphd`

filter. You can also initialize a
`gmphd`

filter used with `trackerPHD`

by
specifying the `FilterInitializationFcn`

property of `trackingSensorConfiguration`

. You can use the provided `initcvgmphd`

, `initctgmphd`

, `initcagmphd`

, and `initctrectgmphd`

as
initialization functions. Or, you can create your own initialization functions.

creates a `phd`

= gmphd`gmphd`

filter with default property values.

allows
you to specify the states and corresponding state covariances of the Gaussian distribution
for each component in the density. `phd`

= gmphd(states,stateCovariances)`states`

and
`stateCovariances`

set the `States`

and `StateCovariances`

properties of the filter.

also allows you to specify properties for the filter using one or more name-value pairs.
Enclose each property name in quotes.`phd`

= gmphd(states,stateCovariances,`Name,Value`

)

`predict` | Predict probability hypothesis density of phd filter |

`correctUndetected` | Correct `phd` filter with no detection hypothesis |

`correct` | Correct `phd` filter with detections |

`likelihood` | Log-likelihood of association between detection cells and components in the density |

`append` | Append two `phd` filter objects |

`merge` | Merge components in the density of `phd` filter |

`scale` | Scale weights of components in the density |

`prune` | Prune the filter by removing selected components |

`labeledDensity` | Keep components with a given label ID |

`extractState` | Extract target state estimates from the `phd` filter |

`clone` | Create duplicate `phd` filter object |

[1] Vo, B. -T, and W. K. Ma. "The
Gaussian mixture Probability Hypothesis Density Filter." *IEEE Transactions on
Signal Processing*, Vol, 54, No, 11, pp. 4091–4104, 2006.

[2] Granstrom, Karl, Christian
Lundquist, and Omut Orguner. "*Extended target tracking using a Gaussian-mixture PHD
filter.*" IEEE Transactions on Aerospace and Electronic Systems 48, no. 4 (2012):
3268-3286.

`trackerPHD`

| `trackingSensorConfiguration`

| `partitionDetections`

| `ggiwphd`

| `initctrectgmphd`

| `initctgmphd`

| `initcvgmphd`

| `initcagmphd`