Figure 1
From: Efficient Closed-loop Maximization of Carbon Nanotube Growth Rate using Bayesian Optimization
An illustration of the closed-loop experimental scheme. On the left is a schematic of the ARES experimental setup, which includes a 532 nm continuous-wave laser, coupled to a high vacuum chamber and a Raman spectrometer. The CNT growth rates are extracted from the collected data (in-situ Raman spectra) and fed into the BO algorithm shown on the right. The algorithm then generates a new set of conditions – the values of the four input variables (\({P}_{1},{P}_{2},{P}_{3},T\)). These experimental conditions are run by ARES, which records the output data and sends it to the BO planner to update the existing dataset and plan a new experiment.
