Learning a Model for Inferring a Spatial Road Lane Network Graph using Self-Supervision

TL;DR

This paper introduces a self-supervised learning approach to generate detailed lane-level road network graphs from onboard sensor data, reducing reliance on costly preconstructed maps and enabling better autonomous vehicle navigation.

Contribution

It presents the first self-supervised method for inferring spatially grounded lane graphs, utilizing a formal model and a novel loss function for robust learning from partial labels.

Findings

Model generalizes to new road layouts

Effective across various intersection types

Outperforms previous approaches in accuracy

Abstract

Interconnected road lanes are a central concept for navigating urban roads. Currently, most autonomous vehicles rely on preconstructed lane maps as designing an algorithmic model is difficult. However, the generation and maintenance of such maps is costly and hinders large-scale adoption of autonomous vehicle technology. This paper presents the first self-supervised learning method to train a model to infer a spatially grounded lane-level road network graph based on a dense segmented representation of the road scene generated from onboard sensors. A formal road lane network model is presented and proves that any structured road scene can be represented by a directed acyclic graph of at most depth three while retaining the notion of intersection regions, and that this is the most compressed representation. The formal model is implemented by a hybrid neural and search-based model, utilizing a novel barrier function loss formulation for robust learning from partial labels. Experiments are conducted for all common road intersection layouts. Results show that the model can generalize to new road layouts, unlike previous approaches, demonstrating its potential for real-world application as a practical learning-based lane-level map generator.