Presentation + Paper
26 October 2018 LetSat-1: a GPU technology demonstrator and AI navigation system
Clifford N. White, Chandler J. Griscom, Prabodh M. Jhaveri, Elizabeth M. Keener, Byron K. Lichtenberg
Author Affiliations +
Abstract
LetSat is a 20-person undergraduate capstone design team at LeTourneau University. LetSat-1 is currently nearing the end of its preliminary design phase. The primary mission of LetSat-1 will be as a technology demonstrator for GPU-based artificial intelligence hardware. The primary component is a NVIDIA Tegra K1 general purpose GPU, with its associated peripherals. To utilize and demonstrate the capabilities of such a GPU, a novel navigation system is being developed. Using machine vision, the AI will analyze downward-looking photographs taken by an onboard camera to determine instantaneous orbital parameters. It is noted that such a scheme will work above any orbital body with permanent terrain, and it could easily be integrated into interplanetary spacecraft.
Conference Presentation
© (2018) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Clifford N. White, Chandler J. Griscom, Prabodh M. Jhaveri, Elizabeth M. Keener, and Byron K. Lichtenberg "LetSat-1: a GPU technology demonstrator and AI navigation system", Proc. SPIE 10769, CubeSats and NanoSats for Remote Sensing II, 107690Y (26 October 2018); https://doi.org/10.1117/12.2319816
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Space operations

Image segmentation

Neural networks

Navigation systems

Image processing

Data processing

Global Positioning System

Back to Top