Hyperdimensional Uncertainty Quantification for Multimodal Uncertainty Fusion in Autonomous Vehicles Perception
Luke Chen, Junyao Wang, Trier Mortlock, Pramod Khargonekar, Mohammad Abdullah Al Faruque
TL;DR
HyperDUM introduces a deterministic, hyperdimensional approach to quantify feature-level epistemic uncertainty in multimodal autonomous vehicle perception. By projecting latent features into hyperdimensional space and forming prototypes, it estimates uncertainty via prototype–feature similarity, with channel-wise (CPB) and patch-wise (PPB) mechanisms preserving per-channel and spatial uncertainty. Pre-fusion uncertainty weighting combined with these prototypes yields consistent improvements in 3D object detection and semantic segmentation under diverse and adversarial conditions, while dramatically reducing FLOPs and parameter counts compared to Bayesian baselines. The method demonstrates robustness to distribution shifts, offers practical efficiency for real-time AV systems, and is supported by comprehensive ablations, cost analyses, and supplemental theory on hyperdimensional expressivity and uncertainty quantifiability.
Abstract
Uncertainty Quantification (UQ) is crucial for ensuring the reliability of machine learning models deployed in real-world autonomous systems. However, existing approaches typically quantify task-level output prediction uncertainty without considering epistemic uncertainty at the multimodal feature fusion level, leading to sub-optimal outcomes. Additionally, popular uncertainty quantification methods, e.g., Bayesian approximations, remain challenging to deploy in practice due to high computational costs in training and inference. In this paper, we propose HyperDUM, a novel deterministic uncertainty method (DUM) that efficiently quantifies feature-level epistemic uncertainty by leveraging hyperdimensional computing. Our method captures the channel and spatial uncertainties through channel and patch -wise projection and bundling techniques respectively. Multimodal sensor features are then adaptively weighted to mitigate uncertainty propagation and improve feature fusion. Our evaluations show that HyperDUM on average outperforms the state-of-the-art (SOTA) algorithms by up to 2.01%/1.27% in 3D Object Detection and up to 1.29% improvement over baselines in semantic segmentation tasks under various types of uncertainties. Notably, HyperDUM requires 2.36x less Floating Point Operations and up to 38.30x less parameters than SOTA methods, providing an efficient solution for real-world autonomous systems.