SafeDrive: Fine-Grained Safety Reasoning for End-to-End Driving in a Sparse World

2026-02-21

AI Summary

End-to-end autonomous driving lacks explicit, interpretable safety reasoning mechanisms, limiting its robustness in complex dynamic environments. To address this, this work proposes SafeDrive, a novel framework that introduces, for the first time, a trajectory-conditioned sparse world model. SafeDrive employs an interaction-aware SWNet to construct a sparse environmental representation and leverages FRNet for fine-grained collision risk assessment and temporal drivable area compliance analysis, enabling explicit modeling of critical traffic participants’ future behaviors and interpretable identification of safety-critical events. Experimental results demonstrate that SafeDrive achieves state-of-the-art performance, attaining 91.6 PDMS and 87.5 EPDMS on NAVSIM with collisions occurring in only 0.5% of scenarios, and a 66.8% driving score on Bench2Drive.

Technology Category

• Natural Language Processing: Safety and Robustness
• Planning, Routing, and Scheduling: Model-Based Reasoning
• Philosophy and Ethics of AI: Safety, Robustness & Trustworthiness

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

The end-to-end (E2E) paradigm, which maps sensor inputs directly to driving decisions, has recently attracted significant attention due to its unified modeling capability and scalability. However, ensuring safety in this unified framework remains one of the most critical challenges. In this work, we propose SafeDrive, an E2E planning framework designed to perform explicit and interpretable safety reasoning through a trajectory-conditioned Sparse World Model. SafeDrive comprises two complementary networks: the Sparse World Network (SWNet) and the Fine-grained Reasoning Network (FRNet). SWNet constructs trajectory-conditioned sparse worlds that simulate the future behaviors of critical dynamic agents and road entities, providing interaction-centric representations for downstream reasoning. FRNet then evaluates agent-specific collision risks and temporal adherence to drivable regions, enabling precise identification of safety-critical events across future timesteps. SafeDrive achieves state-of-the-art performance on both open-loop and closed-loop benchmarks. On NAVSIM, it records a PDMS of 91.6 and an EPDMS of 87.5, with only 61 collisions out of 12,146 scenarios (0.5%). On Bench2Drive, SafeDrive attains a 66.8% driving score.

Problem

• end-to-end driving
• safety reasoning
• autonomous driving
• collision risk
• trajectory planning

Innovation

• end-to-end driving
• safety reasoning
• sparse world model
• trajectory-conditioned representation
• fine-grained risk assessment