Work

J∇TO: Joint Nabla Tracking Optimizer

Year: 2025

Type: article

Abstract: J∇TO (Joint Nabla Tracking Optimizer) is a unified framework for multi-target tracking that integrates statistical gating via Mahalanobis distance, rigid association (Hungarian) via negative likelihood cost, joint probabilistic association (JPDA) with JPDAF update, unified association operator Aτ interpolating Hungarian↔JPDA with anti-switch temporal prior, IMM–Kalman filter, and probabilistic M/N management. It includes statistical consistency, clutter modeling, out-of-order measurements (OOSM), and online calibration of λc and λB. The result is a stable and mathematically coherent framework applicable to highly complex multi-sensor tracking systems.

Source: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.17305454

Author: Jato Braga, Renan

Institution: Massachusetts Institute of Technology

Language: English

Topic: Target Tracking and Data Fusion in Sensor Networks

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W7086947897

FWCI:

Cites:

Cited by:

Related to: