Visualization and Analytics

Multi-object theater plots, detection and object tracks, and track metrics

Use theaterPlot to plot ground truth, coverages, detections, and tracks. Get error metrics for tracks using trackErrorMetrics. Analyze and compare performance of multi-object tracking systems using trackAssignmentMetrics, trackOSPAMetric, and trackGOSPAMetric.

Functions

expand all

Visualization

`theaterPlot`	Plot objects, detections, and tracks in trackingScenario
`coveragePlotter`	Create coverage plotter
`detectionPlotter`	Create detection plotter
`orientationPlotter`	Create orientation plotter
`platformPlotter`	Create platform plotter
`trackPlotter`	Create track plotter
`trajectoryPlotter`	Create trajectory plotter
`plotCoverage`	Plot set of coverages in theater coverage plotter
`plotDetection`	Plot set of detections in theater detection plotter
`plotOrientation`	Plot set of orientations in orientation plotter
`plotPlatform`	Plot set of platforms in platform plotter
`plotTrack`	Plot set of tracks in theater track plotter
`plotTrajectory`	Plot set of trajectories in trajectory plotter
`findPlotter`	Return array of plotters associated with theater plot
`clearData`	Clear data from specific plotter of theater plot
`clearPlotterData`	Clear plotter data from theater plot
`timescope`	Display time-domain signals

Analytics

`trackAssignmentMetrics`	Track establishment, maintenance, and deletion metrics
`trackErrorMetrics`	Track error and NEES
`trackOSPAMetric`	Optimal subpattern assignment (OSPA) metric
`trackGOSPAMetric`	Generalized optimal subpattern assignment (GOSPA) metric
`allanvar`	Allan variance
`magcal`	Magnetometer calibration coefficients

Topics

Configure Time Scope MATLAB Object

Customize timescope properties and use measurement tools.

Featured Examples

Introduction to Tracking Metrics

While designing a multi-object tracking system, it is essential to devise a method to evaluate its performance against the available ground truth. This ground truth is typically available from a simulation environment or by using techniques like ground-truth extraction using manual or automated labeling on recorded data. Though it is possible to qualitatively evaluate a tracking algorithm using visualization tools, the approach is usually not scalable. This example introduces different quantitative analysis tools in Sensor Fusion and Tracking Toolbox™ for assessing a tracker's performance. You will also use some common events like false tracks, track swaps etc. encountered while tracking multiple objects to understand the strengths and limitations of these tools.

Open Live Script

Inertial Sensor Noise Analysis Using Allan Variance

Use the Allan variance to determine noise parameters of a MEMS gyroscope. These parameters can be used to model the gyroscope in simulation. The gyroscope measurement is modeled as:

Open Script

Magnetometer Calibration

Magnetometers detect magnetic field strength along a sensor's X,Y and Z axes. Accurate magnetic field measurements are essential for sensor fusion and the determination of heading and orientation.