Guest Editors Michael T. Eismann and Kenny Chen introduce the Special Section on Application of Artificial Intelligence/Machine Learning to Infrared Systems.
For decades, space grade hardware has always been the determining factor in what we can truly design, develop, and deploy for space missions. Space grade hardware not only has to withstand the harsh elements of space but also needs to be low Size, Weight and Power (SWaP). Those guidelines and requirements dictate the types of techniques, methods, and approaches that can be developed for space applications. Space grade hardware is very expensive and limited in the evaluation kits that are created which leads to the need for a low-cost path to flight development unit for space missions. In recent years Artificial Intelligence and Machine Learning (AI/ML) has skyrocketed in the advancements of what can truly be accomplished. AI/ML is being used to autonomously drive vehicles, track wildfires and help cure diseases. The algorithms created need high performance computing to execute and flourish. In this paper, we present a hardware in the loop implementation for trajectory generation on an AI/ML path to flight development unit. The AI/ML path to flight development unit allows for a common architecture and infrastructure to be used when researching, designing, and developing onboard space AI/ML applications. To aide in the synergy, the Space Common Compute Plug-in Interface (SCCPI) is also presented as a common design interface that allows the deployment of AI/ML applications on space grade hardware. Evaluation of the AI/ML path to flight development unit and leveraging SCCPI is done with several orbital propagation problems running in a variety of power profiles designed to mimic small/cube satellites, spacecraft, and instruments.
We demonstrate a silicon nitride photonics-based imaging system that can perform one-dimensional interferometric imaging around the 1550-nm wavelength. The magnetograph using interferometric and computational imaging for remote observations (MICRO) design uses silicon nitride on a Si platform to replace the bulky free-space optics of traditional magnetograph imaging systems with nanofabricated structures of a fraction of the size. The photonic integrated circuit (PIC) uses an array of lenslets that couple light into four input waveguides with spacing arranged along a Golomb ruler, where each aperture pair formed has a unique length. Each aperture is mixed with a 13-dBm reference laser and separated inside a 2 & times;4 optical hybrid to generate in-phase and quadrature-phase signals to be detected in balanced detectors at the output of the PIC. We use a field programmable gate array (FPGA) board to digitize and process the measurements. The FPGAs and PIC are combined to reduce the overall size, weight, and power of the system, paving the way for a compact imaging system. We demonstrate a PIC-based imager design and experimental testbed for spectrometry applications.
The seldom mentioned ray angle diagram (RAD) allows visualization of important properties of acoustic propagation. Based on the sound speed profile and Snell’s law, the RAD presents depth on the ordinate and tan θ(z; cn) on the abscissa for the full ray cycle of selected rays, parameterized by the ray parameter, cn. The angle θ(z; cn) is the angle the ray makes with the horizontal at depth z. Equivalently, it is the angle of tilt of the wavefront at depth z. Because many useful characteristics of acoustic propagation depend on tan θ(z; cn), it is used as the abscissa of the RAD. The tilt integral is defined as the integral of tan θ(z; cn) with respect to z between the upper and lower limits of the ray. The tilt integral can be visualized by an “area” on the RAD. The WKB phase integral, which relates rays and normal modes, is an integral of tan θ(z; cn) with respect to depth. Individual mode characteristics and number of propagating normal modes are represented on the RAD. Relationships to the adiabatic invariant, time delay for matched field processing, depth dependence of ambient noise, range-averaged transmission loss, and low frequency mode cut-off are discussed.
