ReAL-AD: Towards Human-Like Reasoning in End-to-End Autonomous Driving

2025-07-15

AI Summary

Existing end-to-end autonomous driving methods rely on sparse and fixed trajectory supervision, limiting their ability to model the hierarchical cognitive processes underlying human driving. To address this, we propose the first vision-language model (VLM)-augmented three-tier cognitive reasoning framework for autonomous driving, comprising a strategic reasoning injector, a tactical reasoning integrator, and a hierarchical trajectory decoder—explicitly embedding human-like cognitive structure into end-to-end models. Our framework enables fully interpretable, end-to-end reasoning spanning contextual understanding, hierarchical decision-making, and fine-grained control. Evaluated across multiple benchmarks, it achieves over 30% improvement in planning accuracy and safety, while significantly enhancing behavioral human-likeness, predictability, and decision rationality.

Technology Category

• Computer Vision: Language and Vision
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

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

Abstract

End-to-end autonomous driving has emerged as a promising approach to unify perception, prediction, and planning within a single framework, reducing information loss and improving adaptability. However, existing methods often rely on fixed and sparse trajectory supervision, limiting their ability to capture the hierarchical reasoning process that human drivers naturally employ. To bridge this gap, we propose ReAL-AD, a Reasoning-Augmented Learning framework that structures decision-making in autonomous driving based on the three-tier human cognitive model: Driving Strategy, Driving Decision, and Driving Operation, where Vision-Language Models (VLMs) are incorporated to enhance situational awareness and structured reasoning across these levels. Specifically, we introduce: (1) the Strategic Reasoning Injector, which formulates high-level driving strategies by interpreting complex traffic contexts from VLM-generated insights; (2) the Tactical Reasoning Integrator, which refines strategic intent into interpretable tactical choices such as lane changes, overtaking, and speed adjustments; and (3) the Hierarchical Trajectory Decoder, which progressively translates tactical decisions into precise control actions for smooth and human-like trajectory execution. Extensive evaluations show that integrating our framework improves planning accuracy and safety by over 30%, making end-to-end autonomous driving more interpretable and aligned with human-like hierarchical reasoning. The project page can be found at: href{https://4dvlab.github.io/project_page/realad}{ exttt{4dvlab.github.io/project_page/realad}}

Problem

• Enhance autonomous driving with human-like hierarchical reasoning
• Improve adaptability by integrating Vision-Language Models (VLMs)
• Bridge gap between fixed trajectory supervision and human cognition

Innovation

• Uses Vision-Language Models for situational awareness
• Implements three-tier human cognitive reasoning model
• Hierarchical trajectory decoder for precise control actions