Poster + Presentation + Paper
13 December 2020 Artificial intelligence agent to assist the telescope operation as a troubleshooter using a voice-user interface
Manuel Huertas, Agustín Núñez
Author Affiliations +
Conference Poster
Abstract
The Gran Telescopio Canarias (GTC) is a 10.4-meter optical-infrared telescope located at the Roque de los Muchachos Observatory on the island of La Palma, Spain. Like any modern telescope, it has a high degree of automation and complexity, being made up of hundreds of subsystems that work in a coordinated way to carry out a scientific observation. Even the control system is at a good level of maturity, on certain occasions, when an unwanted event occurs, it sometimes becomes difficult to identify its cause and thus restore the system to its nominal state efficiently. There are three pillars that can be worked on to reduce the time lost due to technical failures: reduce the frequency of occurrence, reduce the impact after the occurrence or increase the detection capacity. However, the system will never be fault-free, so it will always be necessary to develop a system that allows a diagnosis to be made. Ideally, the diagnosis should be made with the information provided by the control system itself and other services that verify the status of the platform, and therefore, with minimal interaction with the user. On the other hand, due to the great variety of faults, and the need to carry out the diagnosis as soon as possible, a voice user interface will be used, in such a way that the operator can indicate the fault that is occurring, using natural language. As a proof of concept, a subsystem diagnostic service has been developed, a thermal monitoring subsystem, a service in charge of providing thermal values of the telescope structure to build a thermal model. During the implementation of this diagnostic service, the need for changes in the software design has already been identified to include the information necessary to generate a good diagnosis, which will indicate to the operator: which part is failing, where it is located and the instruction to replace it. Consequently, it has been proven that extending it to the rest of the control system will bring immediate benefits.
Conference Presentation
© (2020) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Manuel Huertas and Agustín Núñez "Artificial intelligence agent to assist the telescope operation as a troubleshooter using a voice-user interface", Proc. SPIE 11452, Software and Cyberinfrastructure for Astronomy VI, 114522G (13 December 2020); https://doi.org/10.1117/12.2560948
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Artificial intelligence

Telescopes

Control systems

Databases

RELATED CONTENT


Back to Top