GuardAD: Safeguarding Autonomous Driving MLLMs via Markovian Safety Logic
Tianyuan Zhang, Peng Yue, Zihao Peng, Jiangfan Liu, Zonghao Ying, Jiakai Wang, Tianlin Li, Jian Yang, Yaodong Yang, Aishan Liu, Xianglong Liu
Why It Matters
What makes this one worth your time
Improving safety in autonomous driving systems is crucial for real-world deployment, and GuardAD offers a novel approach to address dynamic traffic scenarios.
GuardAD enhances autonomous driving safety using Markovian logic to dynamically refine actions.
Summary
The paper introduces GuardAD, a model-agnostic safeguard for autonomous driving systems that uses Markovian safety logic to enhance safety by continuously updating safety predicates and refining actions based on evolving traffic interactions.
Key contributions
- Introduction of a model-agnostic safeguard using Markovian safety logic.
- Development of Neuro-Symbolic Logic Formalization for safety predicates.
- Demonstration of reduced accident rates and improved task performance in experiments.
Notable insights
- The use of n-th order Markovian Logic Induction allows for the inference of hazards beyond single-step observations.
- Logic-Driven Action Revision refines actions without altering the underlying MLLM, maintaining system integrity.
Possible limitations
- Not stated in the abstract
Abstract
arXiv:2605.10386v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) are increasingly integrated into autonomous driving (AD) systems; however, they remain vulnerable to diverse safety threats, particularly in accident-prone scenarios. Recent safeguard mechanisms have shown promise by incorporating logical constraints, yet most rely on static formulations that lack temporally grounded safety reasoning over evolving traffic interactions, resulting in limited robustness in dynamic driving environments. To address these limitations, we propose GuardAD, a model-agnostic safeguard that formulates AD safety as an evolving Markovian logical state. GuardAD introduces Neuro-Symbolic Logic Formalization, which represents safety predicates over heterogeneous traffic participants and continuously induces them via n-th order Markovian Logic Induction. This design enables the inference of emerging and latent hazards beyond single-step observations. Rather than simply vetoing unsafe actions, GuardAD performs Logic-Driven Action Revision, where inferred safety states actively guide action refinement without modifying the underlying MLLM. Extensive experiments on multiple benchmarks and AD-MLLMs demonstrate that GuardAD substantially reduces accident rates (-32.07%) while slightly improving task performance (+6.85%). Moreover, closed-loop simulation evaluations, together with physical-world vehicle studies, further validate the effectiveness and potential of GuardAD.