Von Mises Based Uncertainty Quantification for Closely Spaced Automotive Radar Targets

2026-06-30

AI Summary

This work addresses the challenge of uncertainty quantification in direction-of-arrival (DOA) estimation for automotive radar in dense target scenarios. It introduces, for the first time, the von Mises distribution from circular statistics into radar DOA modeling and proposes an ensemble method (ENS) based on this distribution. The ENS outputs uncertainty parameters (μ, κ) that respect directional geometric consistency and enable closed-form likelihood computation, allowing seamless integration into downstream data association modules. Compared to evidential deep learning (EDL), ENS yields lower uncertainty under in-distribution conditions and exhibits greater sensitivity to severe perturbations, whereas EDL demonstrates smoother uncertainty transitions and slightly better ranking consistency, revealing a fundamental trade-off between geometric fidelity and statistical generality in uncertainty modeling.

Abstract

This work investigates uncertainty-aware deep learning approaches for direction of arrival (DOA) estimation in automotive radar, focusing on probabilistic modeling and downstream integration. A circular-statistics-based von Mises (VM) ensemble (ENS) is compared with an evidential deep learning (EDL) framework based on a normal inverse gamma formulation, yielding a Student t predictive distribution in the Euclidean domain. The ENS framework produces angular predictions parameterized by (mu, kappa), enabling interpretable uncertainty aligned with directional geometry. Performance is evaluated under in distribution and multiple out-of-distribution conditions using risk coverage and ROC or AUROC analyses. Results indicate that ENS achieves lower uncertainty under nominal conditions and exhibits stronger sensitivity to severe perturbations, whereas EDL provides smoother uncertainty variation and slightly improved ranking consistency. Importantly, the ENS representation enables direct probabilistic integration into association modules via closed form VM likelihoods, facilitating a unified detection tracking pipeline. These findings highlight a trade-off between geometric consistency and statistical generality in uncertainty-aware DOA estimation.

Problem

• uncertainty quantification
• direction of arrival
• automotive radar
• closely spaced targets
• probabilistic modeling

Innovation

• von Mises distribution
• uncertainty quantification
• direction of arrival estimation
• circular statistics
• evidential deep learning