IMMNet: Hybrid Fusion of Model-based and Data-driven Approaches for Maneuvering Target Tracking

2026-07-15

AI Summary

This work addresses the challenges of tracking three-dimensional maneuvering targets, which involve complex dynamics and model mismatch. The authors propose a hybrid approach that integrates the Interacting Multiple Model (IMM) framework with learnable neural components. By preserving the Bayesian inference structure to ensure interpretability while incorporating a data-driven mechanism to adaptively learn target motion patterns and noise characteristics, the method achieves a balance between theoretical rigor and empirical flexibility. The resulting architecture is end-to-end trainable, structurally transparent, and designed to maintain real-time performance, robustness, and tracking accuracy. Experimental results demonstrate that the proposed method significantly outperforms existing algorithms across diverse scenarios, confirming its effectiveness and practical applicability.

Abstract

Maneuvering target tracking in three-dimensional space remains a challenging problem due to complex motion dynamics and model mismatch. To address this, this paper proposes a hybrid model/data-driven algorithm named IMMNet, which integrates the interpretable structure of the interacting multiple model (IMM) algorithm with learnable neural components. Unlike end-to-end black-box methods, the proposed IMMNet algorithm not only can preserve the Bayesian inference mechanism that is essential for real-time radar applications, but also can adaptively learn motion patterns and noise characteristics from data. Extensive experiments demonstrate that the proposed IMMNet algorithm consistently outperforms the existing algorithms across various scenarios, validating it as a robust, interpretable, and practical solution for maneuvering target tracking.

Problem

• maneuvering target tracking
• model mismatch
• motion dynamics
• three-dimensional tracking

Innovation

• hybrid fusion
• maneuvering target tracking
• IMMNet
• model-based and data-driven
• Bayesian inference