arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.01784cs.CVcs.LGstat.ML
收藏

基于自我监督的空间道路网络图推理模型的学习

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

Tier IV Inc.(Tier IV公司);Institute of Innovation for Future Society, Nagoya University(名古屋大学未来社会创新研究所);Graduate School of Informatics, Nagoya University(名古屋大学信息学研究科)

Robin Karlsson, David Robert Wong, Simon Thompson, Kazuya Takeda

摘要:

相互连接的道路车道是城市道路导航的核心概念。目前,大多数自动驾驶车辆依赖于预先构建的车道地图,因为设计算法模型很困难。然而,这种地图的生成和维护成本高昂,阻碍了自动驾驶汽车技术的大规模采用。本文提出了第一个自监督学习方法来训练一个模型来推断一个空间接地车道级道路网络图的基础上,从车载传感器产生的道路场景的密集分段表示。一个正式的道路车道网络模型,并证明了任何结构化的道路场景可以表示为一个有向无环图的深度最多为3,同时保留交叉区域的概念,这是最压缩的表示。形式化模型是由一个混合神经和搜索为基础的模型,利用一种新的障碍函数损失制定的鲁棒学习部分标签。对所有常见的道路交叉口布局进行了实验。结果表明,该模型可以推广到新的道路布局,与以前的方法不同,展示了其作为一个实际的基于学习的车道级地图生成器的现实世界中的应用潜力。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑