Topo2Seq: Enhanced Topology Reasoning via Topology Sequence Learning

2025-02-13

AI Summary

In autonomous driving without high-definition maps, challenges persist in lane topology modeling, long-range perception, and severe misalignment of lane segment endpoints. To address these, this paper proposes an end-to-end monocular image-based lane topology extraction method. Our approach introduces three key contributions: (1) a novel topology sequence learning paradigm that employs randomized ordinal prompts to guide sequence decoding, implicitly encoding the directed acyclic graph (DAG) structure of lane graphs with zero inference overhead; (2) a dual-decoder architecture coupled with joint geometric-topological learning, enabling synergistic optimization under both structural and geometric constraints; and (3) a DAG-aware topology serialization strategy. Evaluated on OpenLane-V2, our method achieves state-of-the-art performance in topology inference, with significant improvements in endpoint alignment accuracy and long-range connection recall.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search

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

Extracting lane topology from perspective views (PV) is crucial for planning and control in autonomous driving. This approach extracts potential drivable trajectories for self-driving vehicles without relying on high-definition (HD) maps. However, the unordered nature and weak long-range perception of the DETR-like framework can result in misaligned segment endpoints and limited topological prediction capabilities. Inspired by the learning of contextual relationships in language models, the connectivity relations in roads can be characterized as explicit topology sequences. In this paper, we introduce Topo2Seq, a novel approach for enhancing topology reasoning via topology sequences learning. The core concept of Topo2Seq is a randomized order prompt-to-sequence learning between lane segment decoder and topology sequence decoder. The dual-decoder branches simultaneously learn the lane topology sequences extracted from the Directed Acyclic Graph (DAG) and the lane graph containing geometric information. Randomized order prompt-to-sequence learning extracts unordered key points from the lane graph predicted by the lane segment decoder, which are then fed into the prompt design of the topology sequence decoder to reconstruct an ordered and complete lane graph. In this way, the lane segment decoder learns powerful long-range perception and accurate topological reasoning from the topology sequence decoder. Notably, topology sequence decoder is only introduced during training and does not affect the inference efficiency. Experimental evaluations on the OpenLane-V2 dataset demonstrate the state-of-the-art performance of Topo2Seq in topology reasoning.

Problem

• Extracting lane topology from perspective views
• Enhancing unordered lane segment alignment
• Improving long-range topological prediction capabilities

Innovation

• Randomized order prompt-to-sequence learning
• Dual-decoder branches for topology sequences
• Training-only topology sequence decoder