MRI: Development of a System for High-Resolution Uninterrupted Capture of Complex Animal Motions

MRI：开发高分辨率不间断捕捉复杂动物运动的系统

• 批准号: 1828280
• 负责人: Rolf Mueller
• 金额: $ 24.97万
• 依托单位: Virginia Polytechnic Institute and State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-10-01 至 2022-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1828280&HistoricalAwards=false
• 关键词: MRI; Development; System; Resolution; Uninterrupted

This major research instrumentation award supports the development of an integrated camera array that will be customized to meet the challenges of capturing fast and highly complex animal motions--providing an enabling tool for fundamental research to understand and model the dynamics of motion. The instrument will permit uninterrupted motion tracking for hundreds of points on an animal as it executes even the most complex motions, such as midair somersaults that bats perform in pursuit of prey. The unprecedented detail, quality, and quantity of the data to be generated will provide the basis for new fundamental research into how animals use freedom of movement to attain unmatched levels of performance in maneuverability and energy efficiency. The large volume of quantitative data produced by the instrument will bring data-intensive methods--from non-linear dynamics and machine learning-- to bear on the field of animal motion. A deeper understanding of the principles behind animal motion will be key to the development of next-generation mobile robots that can handle unconstrained, natural environments. These highly dexterous, mobile robots will enhance productivity in applications such as manufacturing, health care, disaster response, precision agriculture, forestry and environmental monitoring and clean-up. This instrument will also enable fundamental research on the motion of man-made structures, such as the complex dynamic motions inherent in flutter in aerodynamic systems. Knowledge gleaned from this instrument will also help veterinarians to diagnose disease and pain from animals' motion patterns. Graduate and undergraduate students will be involved in instrumentation development and the instrumentation will enable interdisciplinary research training opportunities in engineering and biology. The instrument will combine high spatial and temporal resolution with the ability to view a moving animal from many different angles at the same time. It will consist of 48 high-speed video cameras that can deliver a 1280x1024-pixel image resolution at 1057-Hz frame rate. High-quality illumination will be provided by 8 specialized lights so that no part of a moving animal will ever be hidden from view. All cameras in the array will be synchronized (precision 10 nanoseconds) and operated automatically to allow for efficient capture of large motion data sets. A recording of 5 seconds, for example, will result in over 250,000 images with 332 Gigabytes of raw data. The project team's automated image processing methods will allow reliable tracking of several hundred landmark points across such large image sets. The instrument and its accompanying suite of software tools will be used for the study of previously unexplained animal motion capabilities such as the highly articulated flight of bats, gliding of snakes, and lizards running on vertical substrates.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

这项重大研究仪器奖支持开发集成相机阵列，该阵列将进行定制，以满足捕捉快速且高度复杂的动物运动的挑战，为基础研究提供支持工具，以理解和建模运动动力学。当动物执行最复杂的运动时，例如蝙蝠追逐猎物时进行的空中翻筋斗，该仪器将允许对动物身上的数百个点进行不间断的运动跟踪。所生成的数据的细节、质量和数量前所未有，将为新的基础研究奠定基础，研究动物如何利用运动自由来获得无与伦比的可操作性和能源效率水平。该仪器产生的大量定量数据将为动物运动领域带来来自非线性动力学和机器学习的数据密集型方法。更深入地了解动物运动背后的原理将是开发能够处理不受约束的自然环境的下一代移动机器人的关键。这些高度灵巧的移动机器人将提高制造、医疗保健、灾难响应、精准农业、林业以及环境监测和清理等应用的生产力。该仪器还将支持对人造结构运动的基础研究，例如空气动力系统中颤振固有的复杂动态运动。从该仪器中收集的知识还将帮助兽医根据动物的运动模式诊断疾病和疼痛。研究生和本科生将参与仪器开发，仪器将为工程和生物学领域的跨学科研究培训提供机会。该仪器将结合高空间和时间分辨率以及同时从多个不同角度观察移动动物的能力。它将由 48 个高速摄像机组成，能够以 1057 Hz 帧速率提供 1280x1024 像素的图像分辨率。 8 个专用灯将提供高质量照明，因此移动动物的任何部位都不会被视线遮挡。阵列中的所有摄像机将同步（精度为 10 纳秒）并自动运行，以有效捕获大型运动数据集。例如，录制 5 秒将生成超过 250,000 张图像以及 332 GB 的原始数据。项目团队的自动图像处理方法将能够可靠地跟踪如此大的图像集中的数百个地标点。该仪器及其配套的软件工具套件将用于研究以前无法解释的动物运动能力，例如蝙蝠的高度铰接飞行、蛇的滑行以及在垂直基底上奔跑的蜥蜴。该奖项反映了 NSF 的法定使命，并已被通过使用基金会的智力优点和更广泛的影响审查标准进行评估，认为值得支持。