Field-Programmable System-on-Chip-Based Control System for Real-Time Distortion Correction in Optical Imaging

Abstract

The digital transition requires real-time control of complex systems with short loop time and low latency in various applications. Field-programmable gate arrays (FP-GAs) are, in principle, capable of complying with this task but demand, on the other hand, a high programming effort. In this article, we propose a field-programmable system on chip (FPSoC) as a hybrid solution of an FPGA and a central processing unit (CPU) on a single monolithic die to combine the strengths of both architectures. An FPSoC-based adaptive optical wavefront correction system is presented as a case study to correct camera images in real time that are distorted by time-varying aberrations. While a short total loop time is achieved by interfacing the camera and a deformable mirror on a low level directly with the FPGA, all computationally nonintensive tasks are implemented on the CPU to keep the flexibility, reusability, and development expense low. The system corrects the optical distortion of water surface waves with up to 3600 control cycles per second and spatially attenuates the distortion up to Zernike polynomial 14 with up to 150 Hz. The FPSoC system enables fast spatiotemporal aberration correction in technical processes and offers a perspective for measuring complex flows through fluctuating interfaces.

Particle Tracking Velocimetry With Dynamic Aberration Correction For 3D Flow Measurements Through Fluctuating Phase Boundaries

“…Because the surface of the droplet can oscillate with several hundred Hertz, the digital controller must have a high control cycle rate. Therefore, the signal processing of the wavefront sensor as well as the control algorithm for the deformable mirror are implemented on an in-house developed electronic platform, that is based on a FPSoC chip (Zynq-7100) [15]. This way, 2100 to 3600 control cycles per second can be achieved (depending on algorithm parameters).…”

Green IoT and Edge AI as Key Technological Enablers for a Sustainable Digital Transition towards a Smart Circular Economy: An Industry 5.0 Use Case

“…In the case of FPGAs, they offer very good performance for executing deterministic tasks, but its programming is not as easy as with microcontrollers, and they require to power the used logic continuously. There are also hybrid solutions that combine the benefits of FPGAs and CPUs, known as Field-Programmable Systems-on-Chips (FPSoCs) [45]. In the case of CPLDs, they can execute tasks faster than FPGAs, but their maximum allowed design complexity is inferior to the one offered by FPGAs.…”

Intensity-Only Mode Decomposition on Multimode Fibers Using a Densely Connected Convolutional Network

“…It should be remembered, that the DenseNet has been trained only with synthetic data and works in an experimental environment successfully. The predicted mode weights can be further used for a robust control [38], which is commonly performed in imaging applications based on a field programmable gate array [10], [39].…”