Multi-target Joint Detection, Tracking and Classification Based on Generalized Bayesian Risk using Radar and ESM sensors

TL;DR

This paper introduces a novel Bayesian risk-based method for joint detection, tracking, and classification of multiple targets using radar and ESM sensors, improving accuracy through a unified framework.

Contribution

It develops a new Bayesian risk formulation and a conditional labeled multi-Bernoulli filter for integrated multi-target detection, tracking, and classification.

Findings

Method outperforms existing approaches in simulations.

Enhanced accuracy in multi-target scenarios.

Effective integration of detection, tracking, and classification.

Abstract

In this paper, a novel approach is proposed for multi-target joint detection, tracking and classification based on the labeled random finite set and generalized Bayesian risk using Radar and ESM sensors. A new Bayesian risk is defined for the labeled random finite set variables involving the costs of multi-target cardinality estimation (detection), state estimation (tracking) and classification. The inter-dependence of detection, tracking and classification is then utilized with the minimum Bayesian risk. Furthermore, the conditional labeled multi-Bernoulli filter is developed to calculate the estimates and costs for different hypotheses and decisions of target classes using attribute and dynamical measurements. Moreover, the performance is analyzed. The effectiveness and superiority of the proposed approach are verified using numerical simulations.