UniUncer: Unified Dynamic Static Uncertainty for End to End Driving

2026-03-08

AI Summary

This work addresses the limitation of existing end-to-end driving systems, which typically model uncertainty only in static maps while neglecting that of dynamic traffic participants, leading to unreliable planning in complex interactive scenarios. To overcome this, the authors propose a lightweight, unified uncertainty framework that jointly models both static and dynamic uncertainties for the first time in end-to-end driving. The approach integrates a Laplacian probabilistic regression head, an uncertainty fusion module, and an adaptive gating mechanism, enabling plug-and-play, low-overhead uncertainty incorporation. Compatible with mainstream backbone architectures, the method reduces average trajectory L2 error by 7% on nuScenes and improves the EPDMS metric by 10.8% on NavSimV2, demonstrating notably enhanced robustness—particularly in highly interactive driving situations.

Technology Category

• Reasoning under Uncertainty: Uncertainty Representations
• Planning, Routing, and Scheduling: Planning under Uncertainty
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

End-to-end (E2E) driving has become a cornerstone of both industry deployment and academic research, offering a single learnable pipeline that maps multi-sensor inputs to actions while avoiding hand-engineered modules. However, the reliability of such pipelines strongly depends on how well they handle uncertainty: sensors are noisy, semantics can be ambiguous, and interaction with other road users is inherently stochastic. Uncertainty also appears in multiple forms: classification vs. localization, and, crucially, in both static map elements and dynamic agents. Existing E2E approaches model only static-map uncertainty, leaving planning vulnerable to overconfident and unreliable inputs. We present UniUncer, the first lightweight, unified uncertainty framework that jointly estimates and uses uncertainty for both static and dynamic scene elements inside an E2E planner. Concretely: (1) we convert deterministic heads to probabilistic Laplace regressors that output per-vertex location and scale for vectorized static and dynamic entities; (2) we introduce an uncertainty-fusion module that encodes these parameters and injects them into object/map queries to form uncertainty-aware queries; and (3) we design an uncertainty-aware gate that adaptively modulates reliance on historical inputs (ego status or temporal perception queries) based on current uncertainty levels. The design adds minimal overhead and drops throughput by only $\sim$0.5 FPS while remaining plug-and-play for common E2E backbones. On nuScenes (open-loop), UniUncer reduces average L2 trajectory error by 7\%. On NavsimV2 (pseudo closed-loop), it improves overall EPDMS by 10.8\% and notable stage two gains in challenging, interaction-heavy scenes. Ablations confirm that dynamic-agent uncertainty and the uncertainty-aware gate are both necessary.

Problem

• end-to-end driving
• uncertainty estimation
• static map elements
• dynamic agents
• autonomous driving

Innovation

• uncertainty estimation
• end-to-end driving
• dynamic-static uncertainty
• probabilistic modeling
• uncertainty-aware planning