In this proposed effort, we propose to develop a Deep Learning Processing Subsystem (DLPS) solution for HPSC system. The DLPS solution can significantly improve the performance and energy efficiency of HPSC system in processing deep learning algorithm. The key innovation of this proposal includes design and development of an low power and high performance deep learning processing system which include: (1) low-power and high performance DLPS hardware ; (2) HPSC-compatible software module to manage DLPS hardware and provide API to application layer; (3) DLPS toolchain to transform deep learning models from popular frameworks such as Keras, TensorFlow, and Caffe; (4) DLPS hardware implementation on space grade Xilinx FPGA platform for fault-tolerance design. Finally, all the proposed techniques will be integrated in a functional prototype to demonstrate the feasibility of proposed architecture.
DLPS solution can be extensively used in HPSC system for future NASA missions, which includes, but not limited to, (1) vision-based algorithms such as extreme terrain landing mission (2) model-based reasoning algorithms in vehicle health management system; (3) high-rate instrument data processing algorithms such as in radar applications, and (4) autonomous mission planning algorithms.
DLPS solution can be extensively used for Non-NASA applications, which includes, but not limited to, (1) embedded computer vision system such as Unmanned Aerial Vehicles (UAVs), remote sensing and many other applications (2) satellites imaging (3) self-driving cars (4) robots (5) assistive devices.