Zhang Li, Zhan Haiyang, Liu Xinyuan, Xing Fei, You Zheng
Department of Precision Instrument, Tsinghua University, Beijing, 100084 China.
State Key Laboratory of Precision Measurement Technology and Instrument, Tsinghua University, Beijing, 100084 China.
Microsyst Nanoeng. 2022 Jul 22;8:83. doi: 10.1038/s41378-022-00388-w. eCollection 2022.
Optical measurement systems suffer from a fundamental tradeoff between the field of view (FOV), the resolution and the update rate. A compound eye has the advantages of a wide FOV, high update rate and high sensitivity to motion, providing inspiration for breaking through the constraint and realizing high-performance optical systems. However, most existing studies on artificial compound eyes are limited by complex structure and low resolution, and they focus on imaging instead of precise measurement. Here, a high-performance lensless compound eye microsystem is developed to realize target motion perception through precise and fast orientation measurement. The microsystem splices multiple sub-FOVs formed by long-focal subeyes, images targets distributed in a panoramic range into a single multiplexing image sensor, and codes the subeye aperture array for distinguishing the targets from different sub-FOVs. A wide-field and high resolution are simultaneously realized in a simple and easy-to-manufacture microelectromechanical system (MEMS) aperture array. Moreover, based on the electronic rolling shutter technique of the image sensor, a hyperframe update rate is achieved by the precise measurement of multiple time-shifted spots of one target. The microsystem achieves an orientation measurement accuracy of 0.0023° (3σ) in the direction and 0.0028° (3σ) in the direction in a cone FOV of 120° with an update rate ~20 times higher than the frame rate. This study provides a promising approach for achieving optical measurements with comprehensive high performance and may have great significance in various applications, such as vision-controlled directional navigation and high-dynamic target tracking, formation and obstacle avoidance of unmanned aerial vehicles.
光学测量系统在视场(FOV)、分辨率和更新速率之间存在着根本性的权衡。复眼具有宽视场、高更新速率以及对运动高灵敏度的优点,为突破这种限制并实现高性能光学系统提供了灵感。然而,现有的大多数关于人造复眼的研究都受到结构复杂和分辨率低的限制,并且它们侧重于成像而非精确测量。在此,开发了一种高性能无透镜复眼微系统,以通过精确且快速的方向测量来实现目标运动感知。该微系统拼接由长焦子眼形成的多个子视场,将分布在全景范围内的目标成像到单个复用图像传感器中,并对子眼光阑阵列进行编码,以区分来自不同子视场的目标。在一个简单且易于制造的微机电系统(MEMS)光阑阵列中同时实现了宽视场和高分辨率。此外,基于图像传感器的电子滚动快门技术,通过对一个目标的多个时间偏移光斑进行精确测量,实现了超帧更新速率。该微系统在120°的圆锥视场中,在 方向上实现了0.0023°(3σ)的方向测量精度,在 方向上实现了0.0028°(3σ)的方向测量精度,更新速率比帧率高约20倍。这项研究为实现具有全面高性能的光学测量提供了一种有前景的方法,并且可能在各种应用中具有重要意义,例如视觉控制的定向导航以及无人机的高动态目标跟踪、编队和避障。