Airborne Tracking of Small Ground and Maritime Targets Under Realistic Conditions

现实条件下空中跟踪小型地面和海上目标

• 批准号: 535810-2018
• 负责人: Kirubarajan, Thia
• 金额: $ 26.23万
• 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=720060
• 关键词: Airborne; Tracking; Small; Ground; Maritime

The monitoring of moving objects or persons on the ground using different types of sensors on one or more airborne platforms has many applications in safety & security, search & rescue, law enforcement, intelligent transportation, shipping and agriculture. The sensors on these platforms could be electro-optical/infrared (EO/IR), laser, radar or acoustic. The moving targets of interest could be vehicles, ships, boats, people or even submarines. The targets of interest could be evolving with different motion characteristics (e.g., constant velocity, maneuvering) under different environmental conditions (e.g., low light, rain, heavy traffic). The primary objectives in these airborne surveillance systems are to 1) detect different types of moving objects of interest, 2) track and classify the detected objects, and 3) analyze their behavior under different environmental conditions using the different sensors and limited computing resources onboard. The combination of different types of platforms, sensors, targets and environmental conditions, poses a challenging multiplatform-multisensor-multitarget tracking problem with many conflicting accuracy vs. speed vs. complexity requirements. The objective of this proposal is to address these conflicts in airborne surveillance systems to improve the state-of-the-art in airborne multiplatform-multisensor-multitarget tracking systems. Motivated by the extensive experience of the industry partner in airborne surveillance and their observations from fielded systems, we propose to develop accurate target tracking and fusion algorithms that factor in real-world challenges (e.g., small target sizes, environmental conditions, limited computational resources). While we aim to solve specific challenges in airborne surveillance systems, the results of this research can be applied to intelligent transportation, smart agriculture and law enforcement as well. In addition, the proposed research will train a number of highly qualified personnel in key areas of importance to Canada.

在一个或多个机载平台上使用不同类型的传感器对地面上的移动物体或人员进行监控在安全保障、搜救、执法、智能交通、航运和农业等领域具有多种应用。这些平台上的传感器可以是光电/红外 (EO/IR)、激光、雷达或声学传感器。感兴趣的移动目标可以是车辆、船舶、船只、人员甚至潜艇。感兴趣的目标可能会在不同的环境条件（例如弱光、下雨、交通繁忙）下以不同的运动特性（例如等速、机动）演变。 这些机载监视系统的主要目标是 1) 检测不同类型的感兴趣的移动物体，2) 跟踪和分类检测到的物体，3) 使用不同的传感器和机载有限的计算资源分析它们在不同环境条件下的行为。不同类型的平台、传感器、目标和环境条件的组合提出了具有挑战性的多平台、多传感器、多目标跟踪问题，其精度、速度和复杂性要求之间存在许多相互冲突的要求。该提案的目的是解决机载监视系统中的这些冲突，以提高机载多平台、多传感器、多目标跟踪系统的最先进水平。受行业合作伙伴在机载监视方面的丰富经验及其对现场系统观察的启发，我们建议开发准确的目标跟踪和融合算法，以考虑现实世界的挑战（例如小目标尺寸、环境条件、有限的计算资源） 。虽然我们的目标是解决机载监视系统的具体挑战，但这项研究的结果也可以应用于智能交通、智能农业和执法。此外，拟议的研究还将在对加拿大重要的关键领域培养一批高素质人才。