对于可扩展的自动驾驶而言，独立于 GPS 的鲁棒基于地图的定位系统是基础。为了实现这种基于地图的定位，在线高精（HD）地图构建在位姿的准确估计中发挥着重要作用。尽管由于有效性，近期在线 HD 地图构建的进展主要集中在矢量化表示上，但它们存在计算成本高和固定参数模型的问题，这限制了可扩展性。为了缓解这些限制，我们提出了一种利用图建模的新型 HD 地图学习框架。该框架旨在学习各种几何形状的构建，从而增强 HD 地图构建的可扩展性。我们的方法通过将地图元素分解为顶点和边，将其表示为实例级图，以促进准确高效的端到端矢量化 HD 地图学习。此外，我们引入了一种使用图神经网络的关联策略，以高效处理各种地图元素的复杂几何形状，同时保持可扩展性。在公开开放数据集上的综合实验表明，我们提出的网络比最先进的模型高出 1.6 mAP。我们进一步展示了我们的方法与最先进的方法相比具有卓越的可扩展性，在长距离配置中实现了 4.8 mAP 的提升。我们的代码可在 https://github.com/juyebshin/InstaGraM 获取。
英文摘要：
For scalable autonomous driving, a robust map-based localization system, independent of GPS, is fundamental. To achieve such map-based localization, online high-definition (HD) map construction plays a significant role in accurate estimation of the pose. Although recent advancements in online HD map construction have predominantly investigated on vectorized representation due to its effectiveness, they suffer from computational cost and fixed parametric model, which limit scalability. To alleviate these limitations, we propose a novel HD map learning framework that leverages graph modeling. This framework is designed to learn the construction of diverse geometric shapes, thereby enhancing the scalability of HD map construction. Our approach involves representing the map elements as an instance-level graph by decomposing them into vertices and edges to facilitate accurate and efficient end-to-end vectorized HD map learning. Furthermore, we introduce an association strategy using a Graph Neural Network to efficiently handle the complex geometry of various map elements, while maintaining scalability. Comprehensive experiments on public open dataset show that our proposed network outperforms state-of-the-art model by $1.6$ mAP. We further showcase the superior scalability of our approach compared to state-of-the-art methods, achieving a $4.8$ mAP improvement in long range configuration. Our code is available at https://github.com/juyebshin/InstaGraM.
Improving Hierarchical Representations of Vectorized HD Maps with Perspective Clues
School of Science and Engineering, The Chinese University of Hong Kong (Shenzhen)（科学与工程学院，香港中文大学（深圳））；Undergraduate School of Artificial Intelligence, Shenzhen Polytechnic University（人工智能本科学院，深圳职业技术大学）
School of Vehicle and Mobility, Tsinghua University（清华大学车辆与出行学院）；DiDi Global（滴滴全球公司）；University of Wisconsin-Madison（威斯康星大学麦迪逊分校）；Institute for AI Industry Research (AIR), Tsinghua University（清华大学人工智能产业研究院）
DTCLMapper: Dual Temporal Consistent Learning for Vectorized HD Map Construction
College of Computer Science and Electronic Engineering, Hunan University（湖南大学计算机科学与电子工程学院）；National Engineering Research Center of Optical Instrumentation, Zhejiang University（浙江大学光学仪器国家工程研究中心）；College of Computer Science, Zhejiang University（浙江大学计算机学院）；Institute for Anthropomatics and Robotics, Karlsruhe Institute of Technology（卡尔斯鲁厄理工学院人形机器人与机器人学研究所）；USC Viterbi School of Engineering, the University of Southern California（南加州大学维特比工程学院）
EAN-MapNet: Efficient Vectorized HD Map Construction with Anchor Neighborhoods
School of Intelligent Systems Engineering, Sun Yat-sen University（中山大学智能工程学院）；Guangdong Provincial Key Laboratory of Intelligent Transport System（广东省智能交通系统重点实验室）；China Nuclear Power Design Co.,Ltd.（中国核电工程有限公司）；State Key Laboratory of Nuclear Power Safety Technology and Equipment, China Nuclear Power Engineering Co., Ltd.（中国核工业集团有限公司核电安全技术与装备国家重点实验室）
Online Vectorized HD Map Construction using Geometry
School of Computer Science and Technology, Beijing Institute of Technology（北京理工大学计算机科学与技术学院）；The Chinese University of Hong Kong（香港中文大学）；Tencent AI Lab（腾讯AI实验室）
Tsinghua University（清华大学）；Massachusetts Institute of Technology（麻省理工学院）；Li Auto（理想汽车）
Constructing HD semantic maps is a central component of autonomous driving. However, traditional pipelines require a vast amount of human efforts and resources in annotating and maintaining the semantics in the map, which limits its scalability. In this paper, we introduce the problem of HD semantic map learning, which dynamically constructs the local semantics based on onboard sensor observations. Meanwhile, we introduce a semantic map learning method, dubbed HDMapNet. HDMapNet encodes image features from surrounding cameras and/or point clouds from LiDAR, and predicts vectorized map elements…
Huazhong University of Science & Technology（华中科技大学）；Horizon Robotics（地平线机器人）
Motion prediction is highly relevant to the perception of dynamic objects and static map elements in the scenarios of autonomous driving. In this work, we propose PIP, the first end-to-end Transformer-based framework which jointly and interactively performs online mapping, object detection and motion prediction. PIP leverages map queries, agent queries and mode queries to encode the instance-wise information of map elements, agents and motion intentions, respectively. Based on the unified query representation, a differentiable multi-task interaction scheme is proposed to exploit the correlatio…
