According to the NASA SBIR topic H6.22 description entitled “Deep Neural Net and Neuromorphic Processors for In-Space Autonomy and Cognition,” this subtopic “specifically focuses on advances in signal and data processing. Neuromorphic processing will enable NASA to meet growing demands for applying artificial intelligence and machine learning algorithms on-board a spacecraft to optimize and automate operations. It has been widely received that the recent success of artificial intelligence (AI) is built on three cornerstones: the advance of algorithms, the acquisition of big data, and the availability of high computing power. To further improve the data processing capability and efficiency, researchers, in general, explore from three orthogonal and complementary aspects: algorithm simplification and compression, computing architectures optimized for specific applications, and novel nano-devices that possess unique electrical properties, e.g., synapse- or neuro-alike behavior. These practices are respectively pursued by research societies of machine learning, computer architecture, and solid-state circuit and device. There lacks thorough and sufficient communications and coordination in between. As an example, the quantization of deep neural network (DNN) models often ignores the physical constraints on nano-devices like resistive memory (ReRAM, aka memristor), whose resistance suffers from different variation levels at different resistance values. The higher resistance level can also minimize the power consumption due to the reduced amplitude of the current participating in the computation. Carefully optimizing the quantization scheme of DNNs can achieve both high computational robustness and low power consumption of the ReRAM-based neuromorphic processor.
Technology proposed addresses NASA's TA1 - Flight Computing objectives, which are to “increase onboard autonomy and enable large-scale data triage to support more capable instruments” and “support reliable onboard processing in extreme environments to enable new exploration missions.” And, the TA1 - Ground Computing objectives are to “support 1,000X larger mission computations to enable high-fidelity simulation and large-scale data analysis” and “demonstrate efficient solution of complex NASA problems through quantum and cognitive computing.”
Fast and efficient processing platforms are crucial to the IT revolution. They are poised to meet the performance needs of many important applications: graphics, financial and scientific modeling, biomonitoring, national security scanning, intelligent transportation, networking, multimedia and wireless infrastructure.