trackFuser
Single-hypothesis track-to-track fuser
Description
The trackFuser
System object™ fuses tracks generated by tracking sensors or trackers and allows you to get
fused tracks from decentralized tracking systems. trackFuser
uses the global
nearest neighbor (GNN) algorithm to maintain a single hypothesis about the objects it tracks.
The input tracks are called source or local
tracks, and the output tracks are called central tracks.
To fuse tracks using this object:
Create the
trackFuser
object and set its properties.Call the object with arguments, as if it were a function.
To learn more about how System objects work, see What Are System Objects?
Creation
Description
creates a track-to-track
fuser that uses the global nearest neighbor (GNN) algorithm to maintain a single
hypothesis about the objects it tracks. fuser
= trackFuser
sets properties for the fuser using one or more name-value pairs. Unspecified properties
have default values. Enclose each property name in single quotes.fuser
= trackFuser(Name,Value
)
Properties
Unless otherwise indicated, properties are nontunable, which means you cannot change their
values after calling the object. Objects lock when you call them, and the
release
function unlocks them.
If a property is tunable, you can change its value at any time.
For more information on changing property values, see System Design in MATLAB Using System Objects.
FuserIndex
— Unique index for track fuser
1
(default) | positive integer
Unique index for the fuser, specified as a positive integer. Use this property to distinguish different fusers in a multiple-fuser environment.
Example:
2
MaxNumCentralTracks
— Maximum number of fused central-level tracks
100
(default) | positive integer
Maximum number of fused central-level tracks that the fuser can maintain, specified as a positive integer.
Example:
200
MaxNumSources
— Maximum number of source configurations
20
(default) | positive integer
Maximum number of source configurations that the fuser can maintain, specified as a positive integer.
Example:
200
SourceConfigurations
— Configurations of source systems
cell array of fuserSourceConfiguration
objects
Configurations of source systems, specified as a cell array of fuserSourceConfiguration
objects. The default value is a
1-by-N cell array of fuserSourceConfiguration
objects, where N is the value of the
MaxNumSources
property. You can specify this property during
creation as a Name-Value pair or specify it after creation.
Data Types: object
Assignment
— Assignment algorithm
'MatchPairs'
(default) | 'Munkres'
| 'Jonker-Volgenant'
| 'Auction'
| 'Custom'
Assignment algorithm, specified as 'MatchPairs'
,
'Munkres'
, 'Jonker-Volgenant'
,
'Auction'
, or 'Custom'
. Munkres is the only
assignment algorithm that guarantees an optimal solution, but it is also the slowest,
especially for large numbers of detections and tracks. The other algorithms do not
guarantee an optimal solution but can be faster for problems with 20 or more tracks and
detections. Use'Custom'
to define your own assignment function and
specify its name in the CustomAssignmentFcn
property.
Data Types: char
CustomAssignmentFcn
— Custom assignment function
function handle
Custom assignment function, specified as a function handle. An assignment function must have the following syntax:
[assignments,unassignedCentral,unassignedLocal] = f(cost,costNonAssignment)
assignmunkres
.
Dependencies
To enable this property, set the Assignment
property to
'Custom'
.
Data Types: function handle
| string
| char
AssignmentThreshold
— Track-to-track assignment threshold
30*[1 Inf]
(default) | positive scalar | 1-by-2 vector of positive values
Track-to-track assignment threshold, specified as a positive scalar or a 1-by-2
vector of
[C1,C2],
where C1 ≤
C2. If specified as a scalar, the specified
value, val, is expanded to [val,
Inf
].
Initially, the fuser executes a coarse estimation for the normalized distance between all the local and central tracks. The fuser only calculates the accurate normalized distance for the local and central combinations whose coarse normalized distance is less than C2. Also, the fuser can only assign a local track to a central track if their accurate normalized distance is less than C1. See Algorithms for an explanation of the normalized distance.
Tips:
Increase the value of C2 if there are combinations of local and central tracks that should be calculated for assignment but are not. Decrease it if the calculation takes too much time.
Increase the value of C1 if there are local tracks that should be assigned to central tracks but are not. Decrease it if there are local tracks that are assigned to central tracks they should not be assigned to (too far away).
StateTransitionFcn
— State transition function
'constvel'
(default) | function handle
State transition function, specified as a function handle. This function calculates the state at time step k based on the state at time step k–1.
If
HasAdditiveProcessNoise
istrue
, the function must use the following syntax:where:x(k) = f(x(k-1),dt)
x(k)
— The (estimated) state at timek
, specified as a vector or a matrix. If specified as a matrix, then each column of the matrix represents one state vector.dt
— The time step for prediction.
If
HasAdditiveProcessNoise
isfalse
, the function must use this syntax:where:x(k) = f(x(k-1),w(k-1),dt)
x(k)
— The (estimated) state at timek
, specified as a vector or a matrix. If specified as a matrix, then each column of the matrix represents one state vector.w(k)
— The process noise at timek
.dt
— The time step for prediction.
Example: @constacc
Data Types: function_handle
| char
| string
StateTransitionJacobianFcn
— Jacobian of state transition function
''
(default) | function handle
Jacobian of the state transition function, specified as a function handle. If not specified, the Jacobian is numerically computed, which may increase processing time and numerical inaccuracy. If specified, the function must support one of these two syntaxes:
If
HasAdditiveProcessNoise
istrue
, the function must use this syntax:where:Jx(k) = statejacobianfcn(x(k),dt)
x(k)
— The (estimated) state at timek
, specified as an M-by-1 vector of real values.dt
— The time step for prediction.Jx(k)
— The Jacobian of the state transition function with respect to the state,df/dx
, evaluated atx(k)
. The Jacobian is returned as an M-by-M matrix.
If
HasAdditiveProcessNoise
isfalse
, the function must use this syntax:where:[Jx(k),Jw(k)] = statejacobianfcn(x(k),w(k),dt)
x(k)
— The (estimated) state at timek
, specified as an M-by-1 vector of real values.w(k)
— The process noise at timek
, specified as a W-by-1 vector of real values.dt
— The time step for prediction.Jx(k)
— The Jacobian of the state transition function with respect to the state,df/dx
, evaluated atx(k)
. The Jacobian is returned as an M-by-M matrix.Jw(k)
— The Jacobian of the state transition function with respect to the process noise,df/dw
, evaluated atx(k)
andw(k)
. The Jacobian is returned as an M-by-W matrix.
Example: @constaccjac
Data Types: function_handle
| char
| string
ProcessNoise
— Process noise covariance
eye(3)
(default) | positive real scalar | positive definite matrix
Process noise covariance, specified as a positive real scalar or a positive definite matrix.
When
HasAdditiveProcessNoise
istrue
, specify the process noise covariance as a positive real scalar or a positive definite M-by-M matrix. M is the dimension of the state vector. When specified as a scalar, the matrix is a multiple of the M-by-M identity matrix.When
HasAdditiveProcessNoise
isfalse
, specify the process noise covariance as a W-by-W matrix. W is the dimension of the process noise vector.
Example: [1.0 0.05; 0.05 2
]
Data Types: single
| double
HasAdditiveProcessNoise
— Model additive process noise
false
(default) | true
Option to model process noise as additive, specified as true
or
false
. When this property is true
, process noise
is added to the state vector. Otherwise, noise is incorporated into the state transition
function.
StateParameters
— Parameters of track state reference frame
struct()
(default) | structure | structure array
Parameters of the track state reference frame, specified as a structure or a
structure array. The fuser passes its StateParameters
property
values to the StateParameters
property of the generated tracks. You
can use these parameters to define the reference frame in which the track is reported or
other desirable attributes of the generated tracks.
For example, you can use the following structure to define a rectangular reference
frame whose origin position is at [10 10 0]
meters and whose origin
velocity is [2 -2 0] meters per second with respect to the scenario frame.
Field Name | Value |
---|---|
Frame | "Rectangular" |
Position | [10 10 0] |
Velocity | [2 -2 0] |
Tunable: Yes
Data Types: struct
ConfirmationThreshold
— Threshold for central track confirmation
[2 3]
(default) | positive integer | 1-by-2 vector of positive integers
Threshold for central track confirmation, specified as a positive integer M, or a 1-by-2 vector of positive integers [M N] with M ≤ N. A central track is confirmed if it is assigned to local tracks at least M times in the last N updates. If specified a positive integer M, the confirmation threshold is expanded to [M,M].
Data Types: single
| double
DeletionThreshold
— Threshold for central track deletion
[5 5]
(default) | positive integer | 1-by-2 vector of positive integers
Threshold for central track deletion, specified as a positive integer P, or a 1-by-2 vector of positive integers [P R] with P ≤ R. A central track is deleted if the track is not assigned to local tracks at least P times in the last R updates. If specified a positive integer P, the confirmation threshold is expanded to [P,P].
Example: [5 6]
Data Types: single
| double
FuseConfirmedOnly
— Fuse only confirmed local tracks
true
(default) | false
Fuse only confirmed local tracks, specified as false
or
true
. Set this property to false
if you want to
fuse all local tracks regardless of their confirmation status.
Data Types: logical
FuseCoasted
— Fuse coasted local tracks
false
(default) | true
Fuse coasted local tracks, specified as true
or
false
. Set this property to true
if you want to
fuse coasted local tracks (IsCoasted
field or property of the
localTracks
input is true
). Set it to
false
if you want to only fuse local tracks that are not
coasted.
Example: true
Data Types: logical
StateFusion
— State fusion algorithm
'Cross'
(default) | 'Intersection'
| 'Custom'
State fusion algorithm, specified as:
'Cross'
— Uses the cross-covariance fusion algorithm'Intersection'
— Uses the covariance intersection fusion algorithm'Custom'
— Allows you to specify a customized fusion function
Use the StateFusionParameters
property to specify additional
parameters used by the state fusion algorithm.
Data Types: char
CustomStateFusionFcn
— Custom state fusion function
''
(default) | function handle
Custom state fusion function, specified as a function handle. The state fusion function must support one of the following syntaxes:
[fusedState,fusedCov] = f(trackState,trackCov) [fusedState,fusedCov] = f(trackState,trackCov,fuseParams)
trackState
is specified as an N-by-M matrix. N is the dimension of the track state, and M is the number of tracks.trackCov
is specified as an N-by-N-M matrix. N is the dimension of the track state, and M is the number of tracks.fuseParams
is optional parameters defined in theStateFusionParameters
property.fusedState
is returned as an N-by-1 vector.fusedCov
is returned as an N-by-N matrix.
Dependencies
To enable this property, set the StateFusion
property to
'Custom'
.
Data Types: function_handle
| char
| string
StateFusionParameters
— Parameters for state fusion function
[]
(default)
Parameters for state fusion function. Depending on the choice of
StateFusion
algorithm, you can specify
StateFusionParameters
as:
If
StateFusion
is'Cross'
, specify it as a scalar in (0,1). Seefusexcov
for more details.If
StateFusion
is'Intersection'
, specify it as'det'
or'trace'
. Seefusecovint
for more details.If
StateFusion
is'Custom'
, you can specify these parameters in any variable type, as long as they match the setup of the optionalfuseParams
input of the custom state fusion function specified in theCustomStateFusionFcn
property.
By default, the property is empty.
NumCentralTracks
— Number of central-level tracks
nonnegative integer
This property is read-only.
Number of central tracks currently maintained by the fuser, returned as a nonnegative integer.
Data Types: double
NumConfirmedCentralTracks
— Number of confirmed central tracks
nonnegative integer
This property is read-only.
Number of confirmed central tracks currently maintained by the fuser, returned as a nonnegative integer.
Data Types: double
Usage
Syntax
Description
returns a list of confirmed tracks from a list of local tracks. Confirmed tracks are
predicted to the update time, confirmedTracks
= fuser(localTracks
,tFusion
)tFusion
.
[
also returns a list of tentative tracks, a list of all tracks, and the analysis
information.confirmedTracks
,tentativeTracks
,allTracks
,analysisInformation
] = fuser(localTracks
,tFusion
)
Input Arguments
localTracks
— Local tracks
array of objectTrack
objects | array of structures
Local tracks, specified as an array of objectTrack
objects, or an array
of structures with field names that match the property names of an
objectTrack
object. Local tracks are tracks generated from trackers
in a source track system.
Tip
To specify an empty objectTrack
object, use
objectTrack.empty()
. To specify an empty structure, use
repmat(toStruct(objectTrack),0,0)
.
Data Types: object
| struct
tFusion
— Update time
scalar
Update time, specified as a scalar. The fuser predicts all central tracks to this time. Units are in seconds.
Data Types: single
| double
Output Arguments
confirmedTracks
— Confirmed tracks
array of objectTrack
objects | array of structures
Confirmed tracks, returned as an array of objectTrack
objects in MATLAB® or as an array of structures in code generation. In code
generation, the field names of the returned structure are identical to the
property names of objectTrack
.
The tracker confirms a track if it satisfies the confirmation threshold specified in the
ConfirmationThreshold
property. In that case,
the IsConfirmed
property of the object or field of the
structure is true
.
Data Types: struct
| object
tentativeTracks
— Tentative tracks
array of objectTrack
objects | array of structures
Tentative tracks, returned as an array of objectTrack
objects in
MATLAB or as an array of structures in code generation. In code generation, the
field names of the returned structure are identical to the property names of
objectTrack
.
A track is tentative if it does not satisfy the confirmation threshold specified in the
ConfirmationThreshold
property. In that case, the
IsConfirmed
property of the object or field of the structure is
false
.
Data Types: struct
| object
allTracks
— All tracks
array of objectTrack
objects | array of structures
All tracks, returned as an array of objectTrack
objects in
MATLAB or as an array of structures in code generation. In code generation, the
field names of the returned structure are identical to the property names of
objectTrack
. allTracks
consists of confirmed
and tentative tracks.
Data Types: struct
| object
analysisInformation
— Additional information for analyzing track updates
structure
Additional information for analyzing track updates, returned as a structure. The fields of this structure are:
Field | Description |
TrackIDsAtStepBeginning | Track IDs when the step began |
CostMatrix | Cost of assignment matrix |
Assignments | Assignments returned from the assignment function |
UnassignedCentralTracks | IDs of unassigned central tracks |
UnassignedLocalTracks | IDs of unassigned local tracks |
NonInitializingLocalTracks | IDs of local tracks that were unassigned but were not used to initialize a central track |
InitiatedCentralTrackIDs | IDs of central tracks initiated during the step |
UpdatedCentralTrackIDs | IDs of central tracks updated during the step |
DeletedTrackIDs | IDs of central tracks deleted during the step |
TrackIDsAtStepEnd | IDs of central tracks when the step ended |
Data Types: struct
Object Functions
To use an object function, specify the
System object as the first input argument. For
example, to release system resources of a System object named obj
, use
this syntax:
release(obj)
Specific to trackFuser
predictTracksToTime | Predict track state |
initializeTrack | Initialize new track |
deleteTrack | Delete existing track |
sourceIndices | Fuser source indices |
exportToSimulink | Export tracker or track fuser to Simulink model |
Examples
Fuse Tracks from Two Sources Using trackFuser
Define two tracking sources: one internal and one external. The SourceIndex
of each source must be unique.
internalSource = fuserSourceConfiguration(1,'IsInternalSource',true); externalSource = fuserSourceConfiguration(2,'IsInternalSource',false);
Create a trackFuser
with FuserIndex
equal to 3. The fuser takes the two sources defined above and uses the 'Cross'
StateFusion
model.
fuser = trackFuser('FuserIndex',3, 'MaxNumSources',2, ... 'SourceConfigurations',{internalSource;externalSource}, ... 'StateFusion','Cross');
Update the fuser with two tracks from the two sources. Use a 3-D constant velocity state, in which the states are given in the order of [x; vx; y; vy; z; vz]. The states of the two tracks are the same, but their covariances are different. For the first track, create a large covariance in the x-axis. For the second track, create a large covariance in the y-axis.
tracks = [objectTrack('SourceIndex',1,'State',[10;0;0;0;0;0], ... 'StateCovariance',diag([100,1000,1,10,1,10])); ... objectTrack('SourceIndex',2,'State',[10;0;0;0;0;0], ... 'StateCovariance',diag([1,10,100,1000,1,10]))];
Fuse the track with fusion time equal to 0.
time = 0; confirmedTracks = fuser(tracks,time);
Obtain the positions and position covariances of the source tracks and confirmed tracks.
positionSelector = [1 0 0 0 0 0; 0 0 1 0 0 0; 0 0 0 0 1 0]; % [x; y; z]
[inputPos,inputCov] = getTrackPositions(tracks,positionSelector);
[outputPos,outputCov] = getTrackPositions(confirmedTracks,positionSelector);
Visualize the results using trackPlotter
.
tPlotter = theaterPlot('XLim',[0, 20],'YLim',[-10, 10],'ZLim',[-10, 10]); tPlotter1 = trackPlotter(tPlotter,'DisplayName','Input Tracks','MarkerEdgeColor','blue'); tPlotter2 = trackPlotter(tPlotter,'DisplayName','Fused Tracks','MarkerEdgeColor','green'); plotTrack(tPlotter1,inputPos,inputCov) plotTrack(tPlotter2,outputPos,outputCov) title('Cross-covariance fusion')
References
[1] Blackman, S. and Popoli, R., 1999. Design and analysis of modern tracking systems(Book). Norwood, MA: Artech House, 1999.
[2] Chong, Chee-Yee, Shozo Mori, William H. Barker, and Kuo-Chu Chang. "Architectures and algorithms for track association and fusion." IEEE Aerospace and Electronic Systems Magazine 15, no. 1 (2000): 5-13.
[3] Tian, Xin, Yaakov Bar-Shalom, D. Choukroun, Y. Oshman, J. Thienel, and M. Idan. "Track-to-Track Fusion Architectures-A Review." In book “Advances in Estimation, Navigation, and Spacecraft Control”. Springer, 2015.
Extended Capabilities
C/C++ Code Generation
Generate C and C++ code using MATLAB® Coder™.
Usage notes and limitations:
See System Objects in MATLAB Code Generation (MATLAB Coder).
You must specify the
MaxNumSources
property during construction. Also, the property is read only for code generation.You must specify the
SourceConfigurations
property for all the sources during construction. Additionally,All elements of
SourceConfigurations
must use the sameLocalToCentralTransformFcn
.All elements of
SourceConfigurations
must use the sameCentralToLocalTransformFcn
.
The input tracks must be a struct array instead of an
objectTrack
object array.The track outputs (all three) are each a struct array instead of an
objectTrack
object array.The
ObjectAttributes
structure for all the input source tracks must be in the same format (same field names and data types).The
StateParameters
structure for all the input source tracks must be in the same format (same field names and data types) as theStateParameters
structure of the track fuser.
Version History
Introduced in R2019b
See Also
Functions
assignmunkres
|assignauction
|assignjv
|fusecovint
|fusexcov
|constvel
|constacc
|singer
|singerjac
|constturn
|ctrv
Objects
External Websites
MATLAB コマンド
次の MATLAB コマンドに対応するリンクがクリックされました。
コマンドを MATLAB コマンド ウィンドウに入力して実行してください。Web ブラウザーは MATLAB コマンドをサポートしていません。
Select a Web Site
Choose a web site to get translated content where available and see local events and offers. Based on your location, we recommend that you select: .
You can also select a web site from the following list:
How to Get Best Site Performance
Select the China site (in Chinese or English) for best site performance. Other MathWorks country sites are not optimized for visits from your location.
Americas
- América Latina (Español)
- Canada (English)
- United States (English)
Europe
- Belgium (English)
- Denmark (English)
- Deutschland (Deutsch)
- España (Español)
- Finland (English)
- France (Français)
- Ireland (English)
- Italia (Italiano)
- Luxembourg (English)
- Netherlands (English)
- Norway (English)
- Österreich (Deutsch)
- Portugal (English)
- Sweden (English)
- Switzerland
- United Kingdom (English)