Sensor Fusion and Tracking Toolbox

Design, simulate, and test multisensor tracking and positioning systems

Sensor Fusion and Tracking Toolbox includes tools for designing, simulating, validating, and deploying systems that fuse data from multiple sensors to maintain situational awareness and localization. Reference examples provide a starting point for multi-object tracking and sensor fusion development for surveillance and autonomous systems, including airborne, spaceborne, ground-based, shipborne, and underwater systems.

You can fuse data from real-world sensors such as active and passive radar, sonar, lidar, EO/IR, IMU, and GPS. To further test your tracking algorithms, you can use the simulation environment and sensor models. The toolbox also includes multi-object trackers and estimation filters for evaluating and validating various fusion architectures using track performance metrics such as OSPA and GOSPA.

For simulation acceleration, rapid prototyping, or deployment the toolbox supports C/C++ code generation.

What Is Sensor Fusion and Tracking Toolbox?

Reference Applications

Simulate and track surveillance, autonomous, and inertial navigation systems.

Image illustrating the simulation of a space debris tracking system with a radar surveillance area and space debris tracks represented by lines passing through the surveillance area and orbiting around the globe.

Surveillance Systems

Autonomous Systems

Pose plot showing an object in 3D space with individual lines representing the X, Y, and Z axes of the object.

Inertial Navigation Systems

Product Highlights

Scenario and Sensor Simulation

Define multiplatform scenarios, then assign motion profiles and attach sensor models to each platform. Simulate these scenarios and dynamically visualize platform trajectories, sensor coverages, and object detections.

Tracking Scenario Designer (4:19)

Documentation | Examples

Mountainous terrain showing the trajectories and tracks of a ground target and drone using colored lines.

Multi-Object Tracking

Use a library of domain-specific target and sensor specification to build a multi-object multisensor tracker. Alternatively, use trackers that integrate filters, data association, and track management and choose from a variety of well-known multi-target algorithms, for example multi-hypothesis and joint probabilistic data association.

Bharat Electronics Develops Multi-Target Tracker for Surveillance Radar

Documentation | Examples

Multi-Sensor Fusion

Explore centralized or decentralized multi-object tracking architectures and evaluate design tradeoffs between track-to-track fusion, central-level tracking, or hybrid tracking architectures for tracking applications.

Introduction to Track-To-Track Fusion

Examples

Tracking Data Importer app showing options for converting data, a table with flight data (including time and position), and a visualization of flight paths over a coastal area.

Data Import and Preparation

Use the Tracking Data Importer app to interactively import and convert real-world or simulated tracking truth data for visualization and analysis.

Documentation

Flight trajectory of an aircraft shown as a white line traversing the Earth’s surface, with its altitude, heading, ground speed, and climb rate specified, alongside radar coverage shown as blue ellipses.

Visualization, Evaluation, and Tuning

Analyze and evaluate the performance of tracking systems against ground truth using a variety of tracking metrics. Visualize ground truth, sensor coverages, detections, and tracks on a map or in a MATLAB figure.

Tuning a Multi-Object Tracker

Documentation | Examples

An arrow pointing from MATLAB code to a computer chip, representing the deployment of algorithms to hardware.

Deployment and Hardware Connectivity

Deploy algorithms to hardware targets by automatically generating C/C++ code from fusion and tracking algorithms. Deploy generated code to low-cost, single-precision processors with limited memory.

Processor-in-the-Loop Verification of JPDA Tracker for Automotive Applications

Examples