Abstract:With the rapid advancements in Large Language Models (LLMs), LLM-based agents have introduced convenient and user-friendly methods for leveraging tools across various domains. In the field of astronomical observation, the construction of new telescopes has significantly increased astronomers' workload. Deploying LLM-powered agents can effectively alleviate this burden and reduce the costs associated with training personnel. Within the Nearby Galaxy Supernovae Survey (NGSS) project, which encompasses eight telescopes across three observation sites, aiming to find the transients from the galaxies in 50 mpc, we have developed the \textbf{StarWhisper Telescope System} to manage the entire observation process. This system automates tasks such as generating observation lists, conducting observations, analyzing data, and providing feedback to the observer. Observation lists are customized for different sites and strategies to ensure comprehensive coverage of celestial objects. After manual verification, these lists are uploaded to the telescopes via the agents in the system, which initiates observations upon neutral language. The observed images are analyzed in real-time, and the transients are promptly communicated to the observer. The agent modifies them into a real-time follow-up observation proposal and send to the Xinglong observatory group chat, then add them to the next-day observation lists. Additionally, the integration of AI agents within the system provides online accessibility, saving astronomers' time and encouraging greater participation from amateur astronomers in the NGSS project.

What problem does this paper attempt to address?

### What problems does this paper attempt to solve? This paper aims to solve the problems of labor - intensive tasks and high training costs in astronomical observations. Specifically, with the construction of new telescopes, the workload of astronomers has increased significantly. To solve this problem, the author has developed the StarWhisper Telescope (SWT) system, an astronomical observation system based on large - language - model (LLM) agents, which is used to automate the entire observation process, including generating observation lists, performing observations, analyzing data, and providing feedback to observers. #### Main problems and solutions: 1. **Labor - intensive tasks**: - **Problem**: In the traditional astronomical observation process, a large amount of human participation is required in the observation planning and data processing stages. For example, generating an observation list involves selecting the most suitable celestial bodies from an input catalog containing more than 100,000 entries and considering the specific sky - observing areas of each site and telescope. In addition, the data processing stage requires manual review of nearly 20 GB of raw image data per telescope per day to detect and confirm transient astronomical events. - **Solution**: By introducing the SWT system and using LLM agents to automatically complete these tasks, the burden on astronomers is reduced. 2. **High training costs**: - **Problem**: Training new observers to master complex observation systems and techniques requires a great deal of time and resources. - **Solution**: The SWT system simplifies user operations through natural - language interaction, reducing the learning cost and enabling more amateur astronomers to participate in observation projects. 3. **Improving observation efficiency**: - **Problem**: In the case of multiple telescopes and multiple observation sites, how to manage and coordinate observation tasks efficiently is a challenge. - **Solution**: The SWT system optimizes the assignment and execution of observation tasks by integrating agent workflows, ensuring the smooth progress of the observation process. 4. **Real - time data analysis and feedback**: - **Problem**: Traditional observation systems have delays in data analysis and feedback, which affect the observation efficiency and response speed. - **Solution**: The SWT system implements a real - time data analysis and feedback mechanism, which can quickly notify observers after observing transient events and make follow - up observation suggestions. In conclusion, through the development of the SWT system, this paper aims to reduce the workload of astronomers, lower training costs, improve observation efficiency, and promote more people to participate in astronomical observation activities through automation and intelligent means.

StarWhisper Telescope: Agent-Based Observation Assistant System to Approach AI Astrophysicist

Interpreting Multi-band Galaxy Observations with Large Language Model-Based Agents

SurveyAgent: A Conversational System for Personalized and Efficient Research Survey

The Automatic Observation Management System of the GWAC Network. I. System Architecture and Workflow

What if LLMs Have Different World Views: Simulating Alien Civilizations with LLM-based Agents

The Tsinghua University-Ma Huateng Telescopes for Survey: Overview and Performance of the System

Enhancing GWOPS Capabilities for Coordinated Multi-Telescope Detection of Gravitational Wave Electromagnetic Counterparts

Agents: An Open-source Framework for Autonomous Language Agents

Spectroscopic Observations of the Star Formation Regions in Nearby Galaxies

ModelScope-Agent: Building Your Customizable Agent System with Open-source Large Language Models

Very Large-Scale Multi-Agent Simulation in AgentScope

Multi-Agent System for Cosmological Parameter Analysis

Enabling real-time multi-messenger astrophysics discoveries with deep learning

Telescopic Network of Zhulong for Orbit Determination and Prediction of Space Objects

The Rise and Potential of Large Language Model Based Agents: A Survey

Application of Deep Learning Methods Combined with Physical Background in Wide Field of View Imaging Atmospheric Cherenkov Telescopes

Next-Level, Robotic Telescope-Based Observing Experiences to Boost STEM Enrollments and Majors on a National Scale: Year 1 Report

S3: Social-network Simulation System with Large Language Model-Empowered Agents

AgentScope: A Flexible yet Robust Multi-Agent Platform

ProAgent: Building Proactive Cooperative Agents with Large Language Models