Innovative methods for road infrastructure digitization

道路基础设施数字化的创新方法

• 批准号: 561109-2020
• 负责人: ElBasyouny, Karim
• 金额: $ 3.64万
• 依托单位: University of Alberta
• 依托单位国家: 加拿大
• 项目类别: Alliance Grants
• 财政年份: 2021
• 资助国家: 加拿大
• 起止时间: 2021-01-01 至 2022-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=721883
• 关键词: Innovative; methods; road; infrastructure; digitization

Connected and Autonomous Vehicles (CAVs) hold the promise to bring about a paradigm shift in transportation safety and mobility. Despite recent advances in the automotive industry/research, a fully autonomous system in complex general road environments has not been fully realized, partially due to the bottlenecks caused by AV technologies. The enormous amount of computations and power required for AVs to operate safely in real-time and at high levels of autonomy is beyond the capabilities of any battery-powered vehicle control system. As such, offloading expensive computation workloads to the infrastructure is an active and trending area of research. Recent efforts and collaborative initiatives between the automotive industry, infrastructure owners, and academia shifted towards supporting CAVs with High Definition Maps and semantic road datasets. This project's ultimate goal is to develop methods to build performance-based semantic layers of high definition Maps. Semantic layers of HD maps contain processed information (e.g., traffic signs, obstacles locations, travel lanes, road edges, speed limits, road curvature, gradient, etc.) that can be directly used by CAVs local computer systems. The outcome of this research will facilitate the testing of several vehicle sensor designs in virtual reality on a digitized replica of existing roads collected using light detection and ranging (LiDAR) scanners. The potential outcomes and benefits to Alberta and Canada of this (and similar) projects cannot be overstated: 1) offer a solution to resolve some of the most significant bottlenecks due to existing CAV technologies (e.g., sensor limitations, dynamic occlusions, etc.); 2) improve workload management of local CAV systems; 3) assist government agencies in testing automakers sensor configurations on existing infrastructure in a virtual environment; 4) help stakeholders make informed decisions regarding smart infrastructure investments; 5) allow Canadians to realize the full potential from CAVs over the next decade; and 6) maintain Alberta and Canada's economic and technological competitiveness.

互联和自动驾驶车辆(CAV)有望带来交通安全和机动性的范式转变。尽管汽车工业/研究最近取得了进展，但在复杂的一般道路环境中还没有完全实现完全自主的系统，部分原因是无人机技术造成的瓶颈。自动驾驶系统安全、实时和高度自主运行所需的巨大计算量和电力，超出了任何电池驱动的车辆控制系统的能力。因此，将昂贵的计算工作负载转移到基础设施是一个活跃且有趋势的研究领域。汽车行业、基础设施所有者和学术界最近的努力和合作倡议转向使用高清晰度地图和语义道路数据集支持CAV。该项目的最终目标是开发方法来构建基于性能的高清地图语义层。高清地图的语义层包含经过处理的信息(例如，交通标志、障碍物位置、行车道、道路边缘、速度限制、道路曲率、坡度等)。可以被骑士队的本地计算机系统直接使用。这项研究的结果将有助于在虚拟现实中测试几种车辆传感器设计，在使用光检测和测距(LiDAR)扫描仪收集的现有道路的数字化复制品上进行测试。该项目(以及类似项目)对艾伯塔省和加拿大的潜在成果和好处怎么强调都不为过：1)提供一个解决方案，解决现有CAV技术带来的一些最重大的瓶颈问题(例如传感器限制、动态闭塞等)；2)改善本地CAV系统的工作负荷管理；3)协助政府机构在虚拟环境中测试汽车制造商的现有基础设施；4)帮助利益攸关方就智能基础设施的投资做出明智的决策；5)让加拿大人在未来十年充分发挥CAV的潜力；6)保持艾伯塔省和加拿大的经济和技术竞争力。