T2SG: Traffic Topology Scene Graph for Topology Reasoning in Autonomous Driving

2024-11-28

arXiv.org

AI Summary

In autonomous driving high-definition map construction, lane topology and traffic signal control logic (e.g., right-turn guidance) are often overlooked, leading to incomplete semantic modeling. To address this, we propose the Traffic Topology Scene Graph (T2SG), the first formal representation explicitly encoding lane geometry, signal-guided relationships, and counterfactual road structures. We design TopoFormer—a one-stage Transformer-based architecture—to generate T2SGs: it employs a geometry-distance-driven Lane Aggregation Layer for robust lane grouping and a causality-aware Counterfactual Intervention Layer to model structural alternatives under hypothetical signal changes. Evaluated on OpenLane-V2, our method achieves state-of-the-art T2SG generation performance and attains an OLScore of 46.3 on downstream topological reasoning. The code and models will be publicly released.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Computer Vision: Visual Reasoning & Symbolic Representations

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• 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

Abstract

Understanding the traffic scenes and then generating high-definition (HD) maps present significant challenges in autonomous driving. In this paper, we defined a novel Traffic Topology Scene Graph, a unified scene graph explicitly modeling the lane, controlled and guided by different road signals (e.g., right turn), and topology relationships among them, which is always ignored by previous high-definition (HD) mapping methods. For the generation of T2SG, we propose TopoFormer, a novel one-stage Topology Scene Graph TransFormer with two newly designed layers. Specifically, TopoFormer incorporates a Lane Aggregation Layer (LAL) that leverages the geometric distance among the centerline of lanes to guide the aggregation of global information. Furthermore, we proposed a Counterfactual Intervention Layer (CIL) to model the reasonable road structure ( e.g., intersection, straight) among lanes under counterfactual intervention. Then the generated T2SG can provide a more accurate and explainable description of the topological structure in traffic scenes. Experimental results demonstrate that TopoFormer outperforms existing methods on the T2SG generation task, and the generated T2SG significantly enhances traffic topology reasoning in downstream tasks, achieving a state-of-the-art performance of 46.3 OLS on the OpenLane-V2 benchmark. We will release our source code and model.

Problem

• Modeling traffic topology relationships for autonomous driving
• Generating accurate HD maps using scene graphs
• Enhancing traffic scene understanding with explainable structures

Innovation

• Traffic Topology Scene Graph for autonomous driving
• TopoFormer with Lane Aggregation Layer
• Counterfactual Intervention Layer for road structure