Probabilistic Data Association Filters (PDAF) - a tracking demo

Version 2.0.0.0 (14.1 KB) by Uri Dubin
This code is a short demo of tracking multiple moving points in clutter and noise.
4.9
(8)
5.4K Downloads
Updated 20 Oct 2016

View License

This code is a demo that implements multiple target tracking in 2 and 3 dimensions. It is inspired by work of Y. Bar-Shalom related to Probabilistic Data Association Filters (PDAF).
Main file "Structure_PDAF_Tracking_Demo.m" does the following:
1. Generates number of points moving on different trajectories.
2. Adds clutter and noisy points.
3. Initializes the tracking structures.
4. Implements multiple target tracking over the time.
5. The tracking supports multiple target initiation, occlusion and loss.
This code could be extended to multiple dimensions, target moving profiles and noise. The tracking algorithm tries to follow the original PDAF algorithm. However, some shortcuts are made. Several parameters control the behavior of the tracking.
You are invited to try and comment me on this code.

Cite As

Uri Dubin (2025). Probabilistic Data Association Filters (PDAF) - a tracking demo (/matlabcentral/fileexchange/34146-probabilistic-data-association-filters-pdaf-a-tracking-demo), MATLAB Central File Exchange. Retrieved .

MATLAB Release Compatibility
Created with R2009b
Compatible with any release
Platform Compatibility
Windows macOS Linux

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

PDAF_2D3D/

Version Published Release Notes
2.0.0.0

New version is extended to 3D
Recently I found this link to the theory (I hope it still works)
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.212.383&rep=rep1&type=pdf.
1.0.0.0