High-speed Estimation and Control using Slow-rate Integrative Image Sensors for Adaptive Optics

该奖项的研究目标是实现使用低速图像传感器(如电荷耦合器件(CCD)或互补金属氧化物半导体(CMOS)阵列)进行反馈的高速自适应光学(AO)系统。由于采样率相对较低，这类传感器目前仅限于低带宽应用。建议的研究将利用图像传感器的综合性质，通过结合多速率系统的估计理论(来自系统论)和模糊到运动的算法(来自图像处理)来克服这一带宽限制。该项目的成果包括：(1)将物体的动态转化为图像传感器测量的空间强度分布的系统理论分析；(2)使用慢速图像传感器的快速、准确和稳健的输出估计和识别策略；(3)用于设计和分析具有图像传感器的系统的前馈和反馈规律的工具；以及(4)高速激光光束整形和成像系统的高速波前像差校正的演示。该项目的成功完成将使具有成本效益的低速图像传感器能够在高速应用中使用，这在目前是不可行的，因为低带宽限制。因此，所获得的关键见解不仅将影响高性能的AO系统，还将影响其他依赖图像传感器反馈的高速系统，如视觉伺服机器人、气动光学湍流测量系统和并行光钳。与行业合作伙伴一起，这项研究的结果将传播给来自不同技术的从业者，包括激光加工和高速扫描晶片检测。参与该项目的研究生和本科生将接受对光学、系统理论和图像处理的独特跨学科理解的培训。该项目还将通过开发嵌入可控乐高积木的无源和有源光学元件，向初中生和高中生介绍令人兴奋的自适应光学领域。

The research objective of this award is to realize high-speed adaptive optics (AO) systems that use slow-rate image sensors, such as charge-coupled device (CCD) or complementary metal?oxide?semiconductor (CMOS) arrays, for feedback. Such sensors are currently limited to only low bandwidth applications due to their relatively slow sampling rates. The proposed research will draw on the integrative nature of the image sensor to overcome this bandwidth limitation by combining estimation theory for multi-rate systems (from systems theory) and blur-to-motion algorithms (from image processing). Deliverables from this project include: (1) a system-theoretic analysis of the translation of the dynamics of an object into the spatial intensity distribution measured by an image sensor, (2) fast-rate, accurate, and robust output estimation and identification strategies using slow-rate image sensors; (3) tools for design and analysis of feedforward and feedback laws for systems with image sensors; and (4) demonstration of high-speed laser beam shaping and high-speed wavefront aberration correction for imaging systems. The successful completion of this project will enable the use of cost-effective slow-rate image sensors in high-speed applications, which is currently infeasible because of low bandwidth restrictions. Key insights obtained will thus impact not only high-performance AO systems but also other high-speed systems that rely on image sensor feedback, such as visual-servo robots, aero-optics turbulence measurement systems, and parallelized optical tweezers. Together with industrial partners, the results of this research will be disseminated to practitioners from diverse technologies, including laser processing and wafer inspection with high speed scanning. Graduate and undergraduate students engaged in this project will be trained for a unique interdisciplinary understanding of optics, systems theory, and image processing. The project will also introduce middle and high school students to the exciting area of adaptive optics through development of passive and active optical components embedded into controllable Lego blocks.