Autonomous multi-sensor integration for seamless indoor/outdoor navigation and mobile sensing

自主多传感器集成，实现无缝室内/室外导航和移动传感

• 批准号: RGPIN-2022-03822
• 负责人: ElRabbany, Ahmed
• 金额: $ 2.62万
• 依托单位: Ryerson University
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2022
• 资助国家: 加拿大
• 起止时间: 2022-01-01 至 2023-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=749333
• 关键词: Autonomous; multi; sensor; integration; seamless

The evolution of navigation and perception sensor technologies in recent years has opened the door for the development of low-cost autonomous systems, including autonomous mobile robots (AMR) and small unmanned aerial systems (UAS). Such low-cost systems can potentially be used in many critical applications, both outdoor and indoor. Outdoor applications include topographic mapping, disaster monitoring, precision agriculture, to name a few. Indoor applications, on the other hand, include manufacturing and distribution facilities, health sector, public safety, among others. The use of autonomous systems in these critical applications, however, requires continuous high-accuracy positioning and attitude (pose) information of the platform, along with environmental sensing and object classification. In addition, some use cases of market verticals such as smart manufacturing, warehousing and supply chain, and transportation require seamless indoor/outdoor operational capabilities. The proposed research presents a number of novel ideas, which collectively lead to the development of next generation algorithms and estimation methodologies for a low-cost autonomous multi-sensor integrated system for precise real-time seamless indoor/outdoor navigation and environmental sensing (NES). The proposed system takes advantage of a number of emerging low-cost sensors, including a multi-constellation global navigation satellite systems (GNSS) sensor, a micro-electro-mechanical system (MEMS)-based inertial sensor, a solid state LiDAR sensor, and digital and event cameras. It will also take advantage of emerging fifth-generation (5G) millimeter wave (mmWave) technology, which can help obtain high-accuracy localization in indoor and GNSS-challenging environments. Various levels of sensor/system integration will be developed for different indoor and outdoor environments using nonlinear filtering, deep learning and hybrid models. A software and a prototype system will be developed, which supports the operations of commercial next generation AMR and small UAS. The developed software can be used as a standalone product or it can be integrated with another system, for example, autonomous vehicle control. This in turn could lead to the development of a smart system that can automate the entire AMR or UAV mission. The proposed system overcomes the limitations of current NES systems and is expected to have significant economic and safety benefits to a number of critical industries. The proposed research will also help train a number of highly qualified personnel (HQP), who will be recruited from diverse backgrounds in a manner that ensures equity and inclusion. The HQP will be working collaboratively on the various components of the research project, including algorithm/software development, system design, data collection and system evaluation. Because of the importance of the proposed research, it is expected that the HQP will have a number of high-tech career opportunities.

近年来，导航和感知传感器技术的发展为低成本自主系统的开发打开了大门，包括自主移动机器人（AMR）和小型无人机系统（UAS）。这种低成本系统可用于许多关键应用，包括室外和室内。户外应用包括地形测绘、灾害监测、精准农业等。另一方面，室内应用包括制造和分销设施、卫生部门、公共安全等。然而，在这些关键应用中使用自主系统需要平台的连续高精度定位和姿态（位姿）信息，以及环境传感和物体分类。此外，智能制造、仓储和供应链以及运输等垂直市场的一些用例需要无缝的室内/室外运营能力。所提出的研究提出了许多新颖的想法，这些想法共同促进了低成本自主多传感器集成系统的下一代算法和估计方法的开发，用于精确实时无缝室内/室外导航和环境传感（NES） 。所提出的系统利用了许多新兴的低成本传感器，包括多星座全球导航卫星系统（GNSS）传感器、基于微机电系统（MEMS）的惯性传感器、固态LiDAR传感器、以及数码相机和事件相机。它还将利用新兴的第五代 (5G) 毫米波 (mmWave) 技术，帮助在室内和具有 GNSS 挑战的环境中获得高精度定位。将使用非线性滤波、深度学习和混合模型，针对不同的室内和室外环境开发不同级别的传感器/系统集成。将开发软件和原型系统，支持商业下一代 AMR 和小型无人机的运行。开发的软件可以作为独立产品使用，也可以与其他系统集成，例如自动驾驶车辆控制。这反过来可能会导致智能系统的开发，该系统可以自动化整个 AMR 或无人机任务。所提出的系统克服了当前 NES 系统的局限性，预计将为许多关键行业带来显着的经济和安全效益。拟议的研究还将有助于培训一批高素质人才（HQP），这些人才将以确保公平和包容的方式从不同背景中招募。总部将在该研究项目的各个组成部分进行协作，包括算法/软件开发、系统设计、数据收集和系统评估。由于拟议研究的重要性，预计总部将拥有许多高科技职业机会。