Maneuvering target tracking in three-dimensional space remains a challenging problem due to complex motion dynamics and model mismatch. To address this, the 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 preserves the Bayesian inference mechanism essential for real-time radar applications while adaptively learning motion patterns and noise characteristics from data. Extensive experiments demonstrate that the proposed IMMNet algorithm consistently outperforms existing algorithms across various scenarios, validating it as a robust, interpretable, and practical solution for maneuvering target tracking.
Blogger's Review: The introduction of IMMNet offers a fresh perspective on tackling the dynamic complexities in maneuvering target tracking. By merging interpretability with learning capabilities, it enhances the algorithm's practicality and accuracy, showcasing significant potential in real-time applications. Its fusion of model-based and data-driven approaches may set a crucial research direction for the future.