[论文解读] A telescope control and scheduling system for the Gravitational-wave Optical Transient Observer

M. Dyer|White Rose eTheses Online (University of Leeds, The University of Sheffield, University of York)|Mar 13, 2020

Astronomy and Astrophysical Research被引用 5

一句话总结

本文介绍了GOTO望远镜控制系统（G-TeCS），这是一个完全自主的软件框架，用于控制多个机器人望远镜，以探测引力波事件的电磁对应体。该系统实现了实时调度、天气监控、硬件自愈功能，并整合了多信使警报——成功完成了系统调试，并通过仿真优化为未来全球部署奠定了基础。

ABSTRACT

The detection of the first electromagnetic counterpart to a gravitational-wave signal in August 2017 marked the start of a new era of multi-messenger astrophysics. An unprecedented number of telescopes around the world were involved in hunting for the source of the signal, and although more gravitational-wave signals have been since detected, no further electromagnetic counterparts have been found. In this thesis, I present my work to help build a telescope dedicated to the hunt for these elusive sources: the Gravitational-wave Optical Transient Observer (GOTO). I detail the creation of the GOTO Telescope Control System, G-TeCS, which includes the software required to control multiple wide-field telescopes on a single robotic mount. G-TeCS also includes software that enables the telescope to complete a sky survey and transient alert follow-up observations completely autonomously, whilst monitoring the weather conditions and automatically fixing any hardware issues that arise. I go on to describe the routines used to determine target priorities, as well as how the all-sky survey grid is defined, how gravitational-wave and other transient alerts are received and processed, and how the optimum follow-up strategies for these events were determined. The GOTO prototype, situated on La Palma, saw first light in June 2017. I detail the work I carried out on the site to help commission the prototype, and how the control software was developed during the commissioning phase. I also analyse the GOTO CCD cameras and optics, building a complete theoretical model of the system to confirm the performance of the prototype. Finally, I describe the results of simulations I carried out predicting the future of the GOTO project, with multiple robotic telescopes on La Palma and in Australia, and how G-TeCS might be modified to operate these telescopes as a single, global observatory.

研究动机与目标

• 为引力波光学暂现体观测者（GOTO）项目开发一个可扩展的、自主的望远镜控制与调度系统。
• 利用机器人望远镜网络，实现对引力波及其他瞬变事件的快速响应。
• 在单一调度框架下统一全天区巡天与瞬变源后续观测操作。
• 通过实时天气监控和自动硬件故障恢复，确保系统的韧性。
• 通过建模与优化，为未来在拉帕尔马和澳大利亚部署全球GOTO网络提供支持。

提出的方法

• 设计并实现G-TeCS，一个模块化软件堆栈，用于控制单台机器人赤道仪上的多台大视场望远镜。
• 集成来自引力波（LVC）、伽马射线（Fermi/Swift）及其他瞬变源设施的实时警报数据。
• 利用GOTO-tiling映射方法构建天区巡天网格，以实现所有观测的一致性观测周期。
• 实现基于优先级的调度器，根据事件类型、定位精度和天区覆盖范围动态分配观测任务。
• 构建自愈系统，持续监控硬件状态，并自动恢复曝光错误或赤道仪故障等异常。
• 使用仿真框架对观测周期、巡天效率和真实条件下的响应时间进行建模。

实验结果

研究问题

• RQ1 如何通过单一控制系统高效管理多台机器人望远镜，同时兼顾全天区巡天与瞬变源后续观测？
• RQ2 何种调度策略可在保持巡天周期的前提下，最大化探测引力波对应体的概率？
• RQ3 如何智能地优先处理来自多个设施（如Fermi和Swift伽马暴触发）的重叠警报，以避免重复观测？
• RQ4 为支持未来在多个站点部署的全球GOTO网络，调度器需要进行哪些修改？
• RQ5 系统如何在长时间观测过程中自主恢复硬件与环境故障？

主要发现

• G-TeCS系统成功实现了2017年6月GOTO望远镜在拉帕尔马的首次开机，并自那时起持续运行。
• 系统在调度、天气监控和自愈方面实现了完全自主，显著减少了观测过程中的人员干预。
• 仿真结果证实，将引力波后续观测整合进全天区巡天周期可提升整体巡天效率，同时不损害覆盖范围。
• 系统表明，优先处理高精度定位警报（如来自Swift的警报）而非广域警报（如来自Fermi的警报），可显著减少重复观测，并加快对高概率区域的响应。
• 通过基于仿真的调优，G-TeCS调度器得到优化，展现出更高的性能和更短的瞬变事件响应时间。
• 代码库正在被泛化以支持更广泛复用，计划以开源形式发布，作为兼容Astropy的Python包，供其他机器人望远镜项目使用。