Abstract

With the rapid advancements in Large Language Models (LLMs), LLM-based agentshave introduced convenient and user-friendly methods for leveraging toolsacross various domains. In the field of astronomical observation, theconstruction of new telescopes has significantly increased astronomers'workload. Deploying LLM-powered agents can effectively alleviate this burdenand reduce the costs associated with training personnel. Within the NearbyGalaxy Supernovae Survey (NGSS) project, which encompasses eight telescopesacross three observation sites, aiming to find the transients from the galaxiesin 50 mpc, we have developed the \textbf{StarWhisper Telescope System} tomanage the entire observation process. This system automates tasks such asgenerating observation lists, conducting observations, analyzing data, andproviding feedback to the observer. Observation lists are customized fordifferent sites and strategies to ensure comprehensive coverage of celestialobjects. After manual verification, these lists are uploaded to the telescopesvia the agents in the system, which initiates observations upon neutrallanguage. The observed images are analyzed in real-time, and the transients arepromptly communicated to the observer. The agent modifies them into a real-timefollow-up observation proposal and send to the Xinglong observatory group chat,then add them to the next-day observation lists. Additionally, the integrationof AI agents within the system provides online accessibility, savingastronomers' time and encouraging greater participation from amateurastronomers in the NGSS project.