A Neuro-Symbolic Framework Combining Inductive and Deductive Reasoning for Autonomous Driving Planning

2026-03-12

AI Summary

This work addresses the limitations of existing end-to-end autonomous driving planners in interpretability and safety under long-tail scenarios by proposing the first deeply integrated neuro-symbolic planning framework. The approach leverages a large language model to dynamically extract scene-specific rules, employs answer set programming for logical arbitration to produce discrete decisions, and combines a differentiable kinematic bicycle model with a neural residual network to generate physically feasible continuous trajectories. This architecture enables logically traceable yet end-to-end learnable trajectory generation. Evaluated on the nuScenes benchmark, the method achieves an L2 error of 0.57 meters, a collision rate of merely 0.075%, and a trajectory prediction consistency (TPC) of 0.47 meters, comprehensively outperforming the current state-of-the-art method, MomAD.

Technology Category

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

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Existing end-to-end autonomous driving models rely heavily on purely data-driven inductive reasoning. This "black-box" nature leads to a lack of interpretability and absolute safety guarantees in complex, long-tail scenarios. To overcome this bottleneck, we propose a novel neuro-symbolic trajectory planning framework that seamlessly integrates rigorous deductive reasoning into end-to-end neural networks. Specifically, our framework utilizes a Large Language Model (LLM) to dynamically extract scene rules and employs an Answer Set Programming (ASP) solver for deterministic logical arbitration, generating safe and traceable discrete driving decisions. To bridge the gap between discrete symbols and continuous trajectories, we introduce a decision-conditioned decoding mechanism that transforms high-level logical decisions into learnable embedding vectors, simultaneously constraining the planning query and the physical initial velocity of a differentiable Kinematic Bicycle Model (KBM). By combining KBM-generated physical baseline trajectories with neural residual corrections, our approach inherently guarantees kinematic feasibility while ensuring a high degree of transparency. On the nuScenes benchmark, our method comprehensively outperforms the state-of-the-art baseline MomAD, reducing the L2 mean error to 0.57 m, decreasing the collision rate to 0.075%, and optimizing trajectory prediction consistency (TPC) to 0.47 m.

Problem

• autonomous driving
• inductive reasoning
• interpretability
• safety guarantees
• long-tail scenarios

Innovation

• Neuro-Symbolic Reasoning
• Inductive-Deductive Integration
• Answer Set Programming (ASP)
• Decision-Conditioned Decoding
• Differentiable Kinematic Bicycle Model