Emiko Ito & Dr. Scott Feister
Abstract
A common issue within the professional High Energy Density laser industry is outdated, single-computer control systems. As digital sensors become ubiquitous, quicker, and more productive, dense compilation of tasks puts immense stress on the main computer system. To demonstrate the use of distributed control systems in a mock laboratory environment, our team collaborated with research scientists at Lawrence Livermore National Laboratory to make a closed feedback control system that we could demonstrate at the Supercomputing Conference. Using RGB LEDs, the EPICS (Experimental Physics and Industrial Control System) software stack for distributed control systems, microcontrollers and minicomputers, additive and subtractive manufacturing, we created a simple color matching game for users to experience. Users played a color matching game with the demonstration, and when they were finished, our team used a camera with a machine learning software that tracks the color of the target and iterates through various brightness levels, becoming quicker and more accurate than the human operators over time. With this project demonstrating the ease of integration that machine learning can create while closing feedback loops within experiments, our team has reduced barriers to entry for high power laser scientists to modernize their control systems.
Details
Session 1
11:15am – 12:30pm
Del Norte Hall
Room D: 1530
HSI-SMART
Leave a Reply