Optimal Layered Resource Management and Data Processing for Threat Detection in Urban Environments

城市环境中威胁检测的最佳分层资源管理和数据处理

• 批准号: 538404-2018
• 负责人: Kirubarajan, Thia
• 金额: $ 6.99万
• 依托单位: McMaster University
• 依托单位国家: 加拿大
• 项目类别: Collaborative Research and Development 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=720526
• 关键词: Optimal; Layered; Resource; Management; Data

Canada must be able to secure and protect the people and infrastructure in its urban environments. To do so, threats such as terrorist events and improvised explosive devices, in urban environments need to be identified early not only to safeguard the citizens and the infrastructure, but also to maintain the citizens' confidence in law and order and their trust in the government. In the case of threats that are mobile, it is also essential to accurately track and predict their kinematic states as they evolve over space and time. To detect such stationary or kinematic threats, sensors or signal sources such as cameras, radars, lidars, communication devices, radio and television stations, and cell phone towers can be used in an urban environment.In some urban scenarios, there might be an abundance of sensors due to ubiquitous cameras and cell phones, which could overload the communication and computational resources, or a dearth of accessible sensors due to the presence of buildings that block signal propagation, which would adversely affect the quality of threat detection and tracking. In either case, it is essential to strike a balance between (i) sensing, communication and computational resource utilization and (ii) the desired timeliness and accuracy of threat detection and tracking. This need to balance motivates the proposed research work to develop algorithms for optimal management of sensor, communication and computational resources as well as algorithms for threat detection and object tracking in urban environments using sensors and processors mounted on stationary or moving platforms. The proposed work will train a number of highly qualified personnel (HQP) in sensor and data processing with application to safety and security. In addition, the trained HQP can be employed and the resulting algorithms deployed in other fields such as autonomous vehicles, intelligent transportation and smart cities, which are of importance to maintain Canada's edge in today's high-tech world.

加拿大必须能够保障和保护城市环境中的人员和基础设施。为此，需要及早识别城市环境中的恐怖事件和简易爆炸装置等威胁，这不仅是为了保护公民和基础设施，也是为了维持公民对法律和秩序的信心以及对政府的信任。对于移动威胁，准确跟踪和预测其随空间和时间演变的运动状态也至关重要。为了检测此类静止或运动威胁，可以在城市环境中使用传感器或信号源，例如摄像头、雷达、激光雷达、通信设备、广播电台和电视台以及手机信号塔。在某些城市场景中，可能存在大量的传感器或信号源。由于无处不在的相机和手机而导致传感器数量过多，这可能会使通信和计算资源超载，或者由于建筑物的存在阻碍信号传播而导致缺乏可访问的传感器，这将对威胁检测和跟踪的质量产生不利影响。无论哪种情况，都必须在 (i) 传感、通信和计算资源利用与 (ii) 威胁检测和跟踪所需的及时性和准确性之间取得平衡。这种平衡的需要激发了拟议的研究工作，以开发用于优化管理传感器、通信和计算资源的算法，以及使用安装在固定或移动平台上的传感器和处理器在城市环境中进行威胁检测和对象跟踪的算法。拟议的工作将培训一批传感器和数据处理及安全保障应用方面的高素质人才（HQP）。此外，训练有素的 HQP 以及由此产生的算法可以部署在自动驾驶汽车、智能交通和智慧城市等其他领域，这对于保持加拿大在当今高科技世界的优势具有重要意义。