NIST is developing machine learning-driven autonomous metrology systems as part of their NIST-on-a-Chip program. These systems aim to remove barriers to adoption of advanced metrology technologies by providing laboratory-grade measurement reliability and accuracy in complex operational environments. The project focuses on integrating machine learning with machine-controlled measurement tools for closed-loop experiment design, execution, and analysis.
The team is particularly interested in quantum (NV diamond) and photonic sensor networks for thermodynamic metrology, including temperature, pressure, and humidity measurements. They have demonstrated the use of machine learning models to compensate for long-term drift in Fiber Bragg grating based temperature sensors and are currently working on machine-controlled NV diamond spectrometer for pressure, force, and dosimetry measurements.
NIST is seeking collaborators and volunteers to join their team in developing these autonomous metrology systems. A patent application has been filed for hysteresis compensation in temperature response of fiber Bragg grating thermometers using dynamic regression.
Keywords: quantum, metrology, machine learning, sensor networks
