主要内容

Multi-Object Tracker

Create and manage tracks of multiple objects

expand all in page
  • Library:

  • Automated Driving Toolbox

  • Multi-Object Tracker block

Description

TheMulti-Object Trackerblock initializes, confirms, predicts, corrects, and deletes the tracks of moving objects. Inputs to the multi-object tracker are detection reports generated byDriving Radar Data GeneratorandVision Detection Generator块。多目标跟踪器接受检测s from multiple sensors and assigns them to tracks using a global nearest neighbor (GNN) criterion. Each detection is assigned to a separate track. If the detection cannot be assigned to any track, the multi-object tracker creates a new track.

A new track starts in atentativestate. If enough detections are assigned to a tentative track, its status changes toconfirmed。When a track is confirmed, the multi-object tracker considers that track to represent a physical object. If detections are not added to the track within a specifiable number of updates, the track is deleted.

The multi-object tracker also estimates the state vector and state vector covariance matrix for each track using a Kalman filter. These state vectors are used to predict a track's location in each frame and determine the likelihood of each detection being assigned to each track.

Ports

Input

expand all

Detection list, specified as a Simulink bus containing a MATLAB structure. SeeGroup Signal Lines into Virtual Buses(Simulink)。The structure has the form:

Field Description Type
NumDetections Number of detections integer
IsValidTime falses当有时在块调用间隔之间要求更新时 Boolean
Detections Object detections 对象检测结构的数组。首先NumDetectionsof these detections are actual detections.

The definitions of the object detection structures are found in theDetectionsoutput port descriptions of theDriving Radar Data GeneratorandVision Detection Generator块。

Note

The object detection structure contains aTimefield. The time tag of each object detection must be less than or equal to the time of the current invocation of the block. The time tag must also be greater than the update time specified in the previous invocation of the block.

Track update time, specified as a real scalar. The multi-object tracker updates all tracks to this time. Update time must always increase with each invocation of the block. Units are in seconds.

Note

The object detection structure contains aTimefield. The time tag of each object detection must be less than or equal to the time of the current invocation of the block. The time tag must also be greater than the update time in the previous invocation of the block.

Dependencies

To enable this port, setPrediction time sourcetoInput port。

Cost matrix, specified as a real-valuedNt-by-Nd矩阵，哪里Ntis the number of existing tracks andNdis the number of current detections.

The rows of the cost matrix correspond to the existing tracks. The columns correspond to the detections. Tracks are ordered as they appear in the list of tracks in theAll Tracksoutput port of the previous invocation of the block.

In the first update to the multi-object tracker, or if the track has no previous tracks, assign the cost matrix a size of [0,Nd]. The cost must be calculated so that lower costs indicate a higher likelihood that the multi-object tracker assigns a detection to a track. To prevent certain detections from being assigned to certain tracks, useinf。

Dependencies

要启用此端口，请选择Enable cost matrix input。

Detectable track IDs, specified as a real-valuedM-b-1矢量或M-by-2 matrix. Detectable tracks are tracks that the sensors expect to detect. The first column of the matrix contains a list of track IDs that the sensors report as detectable. The optional second column contains the detection probability for the track.

Tracks whose identifiers are not included inDetectable Track IDsare considered undetectable. The track deletion logic does not count the lack of detection as a "missed detection" for track deletion purposes.

If this port is not enabled, the tracker assumes all tracks to be detectable at each invocation of the block.

Dependencies

To enable this port, in thePort Settingtab, selectEnable detectable track IDs Input。

跟踪状态参数, specified as a Simulink bus containing a MATLAB structure. The structure has the form:

Field Description
NumParameters Number of non-default state parameters, specified as a nonnegative integer
Parameters Array of state parameter structures

The block uses the value of theParametersfield for theStateParametersfield 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]仪表，其起源速度是[2 -2 0]meters per second with respect to the scenario frame.

字段名称 Value
Frame "Rectangular"
Position [10 10 0]
速度 [2 -2 0]

Dependencies

To enable this port, in the跟踪器配置tab, select theUpdate track state parameters with timeparameter.

Output

expand all

Confirmed tracks, returned as a Simulink bus containing a MATLAB structure. SeeCreate Nonvirtual Buses(Simulink)。

This table shows the structure fields.

Field Description
NumTracks Number of tracks
Tracks Array of track structures of a length set by the最大轨道数量parameter. Only the firstNumTracksof these are actual tracks.

This table shows the fields of each track structure.

Field Definition
TrackID Unique track identifier used to distinguish multiple tracks.
BranchID Unique track branch identifier used to distinguish multiple track branches.
SourceIndex Unique source index used to distinguish tracking sources in a multiple tracker environment.
UpdateTime 曲目更新的时间。单位在几秒钟内。
Age 赛道更新的次数。
State

Value of state vector at the update time.
StateCovariance

Uncertainty covariance matrix.
ObjectClassID 代表对象分类的整数值。价值0代表未知分类。非零分类仅适用于已确认的轨道。
TrackLogic Confirmation and deletion logic type. This value is always'History'for radar sensors, to indicate history-based logic.
TrackLogicState

轨道逻辑类型的当前状态，返回为1乘Klogical array.Kis the number of latest track logical states recorded. In the array,1denotes a hit and0denotes a miss.
IsConfirmed Confirmation status. This field istrueif the track is confirmed to be a real target.
IsCoasted Coasting status. This field istrueif the track is updated without a new detection.
IsSelfReported

指示跟踪器是否报告了轨道。该字段用于轨道融合环境中。它被返回为trueby default.
objectattributes Additional information about the track.

For more details about these fields, seeobjectTrack。

如果以下方式确认轨道

  • At leastMdetections are assigned to the track during the firstNupdates after track initialization. To specify the valuesMandN, use theM and N for the M-out-of-N confirmationparameter.

  • The detection initiating the track has anObjectClassIDgreater than zero.

Tentative tracks, returned as a Simulink bus containing a MATLAB structure. SeeCreate Nonvirtual Buses(Simulink)。A track is tentative before it is confirmed.

This table shows the structure fields.

Field Description
NumTracks Number of tracks
Tracks Array of track structures of a length set by the最大轨道数量parameter. Only the firstNumTracksof these are actual tracks.

This table shows the fields of each track structure.

Field Definition
TrackID Unique track identifier used to distinguish multiple tracks.
BranchID Unique track branch identifier used to distinguish multiple track branches.
SourceIndex Unique source index used to distinguish tracking sources in a multiple tracker environment.
UpdateTime 曲目更新的时间。单位在几秒钟内。
Age 赛道更新的次数。
State

Value of state vector at the update time.
StateCovariance

Uncertainty covariance matrix.
ObjectClassID 代表对象分类的整数值。价值0代表未知分类。非零分类仅适用于已确认的轨道。
TrackLogic Confirmation and deletion logic type. This value is always'History'for radar sensors, to indicate history-based logic.
TrackLogicState

轨道逻辑类型的当前状态，返回为1乘Klogical array.Kis the number of latest track logical states recorded. In the array,1denotes a hit and0denotes a miss.
IsConfirmed Confirmation status. This field istrueif the track is confirmed to be a real target.
IsCoasted Coasting status. This field istrueif the track is updated without a new detection.
IsSelfReported

指示跟踪器是否报告了轨道。该字段用于轨道融合环境中。它被返回为trueby default.
objectattributes Additional information about the track.

For more details about these fields, seeobjectTrack。

Dependencies

要启用此端口，请选择Enable tentative tracks output。

Combined list of confirmed and tentative tracks, returned as a Simulink bus containing a MATLAB structure. SeeCreate Nonvirtual Buses(Simulink)。

This table shows the structure fields.

Field Description
NumTracks Number of tracks
Tracks Array of track structures of a length set by the最大轨道数量parameter. Only the firstNumTracksof these are actual tracks.

This table shows the fields of each track structure.

Field Definition
TrackID Unique track identifier used to distinguish multiple tracks.
BranchID Unique track branch identifier used to distinguish multiple track branches.
SourceIndex Unique source index used to distinguish tracking sources in a multiple tracker environment.
UpdateTime 曲目更新的时间。单位在几秒钟内。
Age 赛道更新的次数。
State

Value of state vector at the update time.
StateCovariance

Uncertainty covariance matrix.
ObjectClassID 代表对象分类的整数值。价值0代表未知分类。非零分类仅适用于已确认的轨道。
TrackLogic Confirmation and deletion logic type. This value is always'History'for radar sensors, to indicate history-based logic.
TrackLogicState

轨道逻辑类型的当前状态，返回为1乘Klogical array.Kis the number of latest track logical states recorded. In the array,1denotes a hit and0denotes a miss.
IsConfirmed Confirmation status. This field istrueif the track is confirmed to be a real target.
IsCoasted Coasting status. This field istrueif the track is updated without a new detection.
IsSelfReported

指示跟踪器是否报告了轨道。该字段用于轨道融合环境中。它被返回为trueby default.
objectattributes Additional information about the track.

For more details about these fields, seeobjectTrack。

Dependencies

要启用此端口，请选择Enable all tracks output。

Parameters

expand all

Tracker Management

Unique tracker identifier, specified as a nonnegative integer. This parameter is used as theSourceIndexin the outputs, and distinguishes tracks that come from different trackers in a multiple-tracker system. You must specify this property as a positive integer to use the track outputs as inputs to a track fuser.

Example:1

Kalman filter initialization function, specified as a function name. The toolbox provides several initialization functions. For an example of an initialization function, seeinitcvekf。

检测分配阈值，指定为正真实标量。要将检测分配到轨道，检测到距轨道的归一化距离必须小于分配阈值。如果某些检测仍然与要分配给的轨道未分配，请增加阈值。如果将某些检测分配给不正确的轨道，请降低阈值。

Confirmation parameters for track creation, specified as a two-element vector of positive integers,[M,N]。A track is confirmed when at leastMdetections are assigned to the track during the firstNupdates after track initialization.Mmust be less than or equal toN。

  • When settingN, consider the number of times you want the tracker to update before it confirms a track. For example, if a tracker updates every 0.05 seconds, and you allow 0.5 seconds to make a confirmation decision, setN= 10.

  • When settingM, take into account the probability of object detection for the sensors. The probability of detection depends on factors such as occlusion or clutter. You can reduceMwhen tracks fail to be confirmed or increaseMwhen too many false detections are assigned to tracks.

Example:[3,5]

Track deletion threshold for history logic, specified as a real-valued 1-by-2 vector of positive integers[P R]。If a confirmed track is not assigned to any detectionPtimes in the lastRtracker updates, then the track is deleted.

最大轨道数量that the block can process, specified as a positive integer.

Maximum number of sensors that the block can process, specified as a positive integer. This value should be greater than or equal to the highestSensorIndexvalue used in theDetectionsinput port.

Out-of-sequence measurements handling, specified asTerminateorneglect。每个检测都有与之相关的时间戳，td, and the tracker block has it own timestamp,tt, which is updated in each invocation. The tracker block considers a measurement as an OOSM iftd<tt。

When the parameter is specified as:

  • Terminate— The block stops running when it encounters any out-of-sequence measurements.

  • Neglect— The block neglects any out-of-sequence measurements and continue to run.

Specify the parameters of the track state reference frame as a structure or a structure array. The block passes the value of this parameter to theStateParametersfield 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]仪表，其起源速度是[2 -2 0]meters per second with respect to the scenario frame.

字段名称 Value
Frame "Rectangular"
Position [10 10 0]
速度 [2 -2 0]

You can update the track state parameters through theState Parametersinput port by selecting theUpdate track state parameters with timeparameter.

Data Types:struct

选择this parameter to enable the input port for track state parameters through theState Parametersinput port.

Inputs and Outputs

Source for prediction time, specified asInput portorAuto。选择Input portto input an update time by using thePrediction Timeinput port. Otherwise, the simulation clock managed by Simulink determines the update time.

Example:Auto

选择此复选框以启用使用成本矩阵的输入Cost Matrixinput port.

选择this check box to enable theDetectable Track IDsinput port.

输出总线名称的来源，指定为AutoorProperty。

  • If you selectAuto, the block automatically creates a bus name.

  • If you selectProperty, specify the bus name using theSpecify an output bus nameparameter.

Dependencies

To enable this parameter, set theSource of output bus nameparameter toProperty。

选择this check box to enable the output of tentative tracks by using theTentative Tracksoutput port.

选择this check box to enable the output of all the tracks by using theAll Tracksoutput port.

  • Interpreted execution— Simulate the model using the MATLAB interpreter. This option shortens startup time. InInterpreted executionmode, you can debug the source code of the block.

  • 代码生成— Simulate the model using generated C/C++ code. The first time you run a simulation, Simulink generates C/C++ code for the block. The C code is reused for subsequent simulations as long as the model does not change. This option requires additional startup time.

Model Examples

Sensor Fusion Using Synthetic Radar and Vision Data in Simulink

Sensor Fusion Using Synthetic Radar and Vision Data in Simulink

Implement a synthetic data simulation for tracking and sensor fusion in Simulink with Automated Driving Toolbox™.
开放型号
Adaptive Cruise Control with Sensor Fusion

Adaptive Cruise Control with Sensor Fusion

Implement an automotive adaptive cruise controller using sensor fusion.
开放型号

Extended Capabilities

Version History

Introduced in R2017b

See Also

Apps

Blocks

Objects