Multi-Sensor Precise Positioning for Autonomous and Connected Vehicles

自动驾驶和联网车辆的多传感器精确定位

• 批准号: RGPIN-2020-03900
• 负责人: Noureldin, Aboelmagd
• 金额: $ 2.84万
• 依托单位: Royal Military College of Canada
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2020
• 资助国家: 加拿大
• 起止时间: 2020-01-01 至 2021-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=716710
• 关键词: Multi; Sensor; Precise; Positioning; Autonomous; Multi; Sensor; Precise; Positioning; Autonomous

Autonomous and connected vehicles (AV/CVs) promise to improve the safety and efficiency of our transportation system and bring new economic opportunities. They have the potential to reduce collisions, traffic congestion and emissions, and to improve mobility for a growing population. The increasing demand for AV/CVs is shaping the future of the automotive industry by connecting vehicles to one another and the surrounding infrastructure, transforming the in-vehicle experience and paving the way to autonomous driving. AV technology requires vehicle onboard intelligence relying on a suite of sensors and systems involving global navigation satellite systems (GNSS) including GPS, vehicle motion sensors (gyroscopes, accelerometers, and speedometers), cameras, light detection and ranging (LiDAR) and radar. On the other hand, CV technology involves communication between vehicles to share a common perception and awareness of other traffic participants. A key AV/CV system requirement is to have a highly reliable positioning to sub-meter lane-level accuracy. The availability of the above-mentioned sensors and systems in future AV/CVs provides an attractive opportunity to further increase the accuracy and reliability of the positioning system. The proposed research program aims to provide continuous, reliable and ultra-precise multi-sensor positioning for the safe operation of AV/CVs. The proposed research will enhance the positioning of AV/CVs on multiple fronts including (a) accurate positioning reaching decimeter levels of accuracy for lane-level navigation with high integrity; (b) advanced multi-sensor fusion for positioning anywhere for autonomy that adapts to changes in the environment; (c) machine intelligence algorithms to select the set of sensors and systems to be integrated to provide the most reliable and accurate positioning information in all operational environments; and (d) high data rate positioning information for fast response to meet the demand for safety-critical autonomous driving. This research program will enable the deployment of novel AV/CV positioning systems operating efficiently everywhere even in degraded and challenging environments such as areas of GNSS signal blockage, darkness, absence of visual features, smoke, snow, fog, and rain while providing decimeter-level accuracy. From an autonomy perspective, this work is expected to have a significant impact on the acceptance and trust of targeted AV/CVs since it will increase reliability, decrease the failure rate expected in such degraded environments, and enhance safety and reliability. This research will also contribute to training highly qualified personnel (HQP) with expertise in multi-sensor fusion, machine intelligence and precise positioning. There is a strong demand for HQP with knowledge in these areas and their future employment will accelerate the translation of next-generation multi-sensory positioning and autonomous technologies to the Canadian industry.

自动驾驶汽车（AV/CVS）有望提高运输系统的安全性和效率，并带来新的经济机会。他们有可能减少碰撞，交通拥堵和排放，并改善人口不断增长的流动性。对AV/CVS的需求不断增长，是通过将车辆彼此连接到彼此和周围的基础设施，改变车辆内的体验并为自动驾驶铺平道路，从而塑造了汽车行业的未来。 AV技术需要依靠一套涉及全球导航卫星系统（GNSS）的传感器和系统的车载智能，包括GPS，车辆运动传感器（陀螺仪，加速度计和速度计），相机，光检测和范围（LIDAR）和雷达。另一方面，简历技术涉及在车辆之间进行沟通，以分享其他交通参与者的共同看法和意识。关键的AV/CV系统要求是对子米车道级别的准确性具有高度可靠的定位。将来的AV/CVS中上述传感器和系统的可用性为进一步提高定位系统的准确性和可靠性提供了一个有吸引力的机会。 拟议的研究计划旨在为AV/CVS的安全操作提供持续，可靠和超专注的多传感器定位。拟议的研究将增强AV/CVS在多个方面的定位，包括（a）具有高完整性的车道级导航的准确定位； （b）高级多传感器融合，用于定位在适应环境变化的自主权的任何地方； （c）机器智能算法选择要集成的传感器和系统集，以在所有操作环境中提供最可靠，最准确的定位信息； （d）快速响应的高数据速率定位信息，以满足对安全至关重要自动驾驶的需求。 该研究计划将使新型的AV/CV定位系统的部署即使在降级和具有挑战性的环境中，例如GNSS信号阻塞，黑暗，视觉特征，烟雾，雪，雾，雾气和雨水，同时提供分解器水平的准确性。从自治的角度来看，这项工作有望对有针对性的AV/CVS的接受和信任产生重大影响，因为它将提高可靠性，降低此类退化环境中预期的失败率，并提高安全性和可靠性。 这项研究还将为具有多传感器融合，机器智能和精确定位方面的专业知识而培训高素质的人员（HQP）。在这些领域的知识中，对HQP的需求很强，未来的就业将加速下一代多感官定位和自治技术向加拿大行业的翻译。