|国家预印本平台
| 注册
首页|Hybrid rack-machine-learning pointing correction for a 1 m robotic telescope

Hybrid rack-machine-learning pointing correction for a 1 m robotic telescope

Weiqing Hu Qingsheng Zhu

✕

Hybrid rack-machine-learning pointing correction for a 1 m robotic telescope

Hybrid rack-machine-learning pointing correction for a 1 m robotic telescope

Weiqing Hu 1Qingsheng Zhu2

作者信息

  • 1. Nanjing Astronomical Instruments Research Center, Chinese Academy of Sciences, Nanjing 210042, China;School of Astronomy and Space Science, University of Science and Technology of China, Hefei 230026, China
  • 2. Nanjing Astronomical Instruments Research Center, Chinese Academy of Sciences, Nanjing 210042, China;CAS Nanjing Astronomical Instruments Co, Ltd, Nanjing 210042, China;School of Astronomy and Space Science, University of Science and Technology of China, Hefei 230026, China
  • 折叠

摘要

We present a hybrid pointing-correction framework for robotic telescopes and evaluate it under strict temporal generalization using real Alt-Az guiding data from a 1 m telescope in the Las Cumbres Observatory network. The framework combines a 19-parameter rack model as a physically motivated analytic backbone with an XGBoost-based residual regressor that learns the remaining error from rack-term contributions, Alt-Az geometry, and a compact set of environmental features. The primary evaluation uses rolling temporal splits, so that each test segment is strictly later than the data used for training and model selection, with a hybrid strategy achieving the best overall future-window pointing accuracy among the tested model families. This shows a clear advantage over the purely analytic and purely data-driven baselines, while the improvement over a simpler residual-learning baseline is more modest. In terms of weighted mean vector root-mean-square residual (VecRMS) the corresponding values are 12.75″ for Hybrid, 14.95″ for Rack+ResTree, 18.14″ for Rack-only, and 21.87″ for Pure Machine Learning (PureML). Additional drift diagnostics provide descriptive evidence of a temporal distribution shift over the studied interval, supporting the use of time-aware evaluation. A strictly causal offline replay also suggests that a conservative validation-gated model-refresh policy may improve future-window pointing accuracy under controlled conditions. Additional analyses of condition-wise robustness, feature relevance, and computational overhead support the practical feasibility of the framework.

Abstract

We present a hybrid pointing-correction framework for robotic telescopes and evaluate it under strict temporal generalization using real Alt-Az guiding data from a 1 m telescope in the Las Cumbres Observatory network. The framework combines a 19-parameter rack model as a physically motivated analytic backbone with an XGBoost-based residual regressor that learns the remaining error from rack-term contributions, Alt-Az geometry, and a compact set of environmental features. The primary evaluation uses rolling temporal splits, so that each test segment is strictly later than the data used for training and model selection, with a hybrid strategy achieving the best overall future-window pointing accuracy among the tested model families. This shows a clear advantage over the purely analytic and purely data-driven baselines, while the improvement over a simpler residual-learning baseline is more modest. In terms of weighted mean vector root-mean-square residual (VecRMS) the corresponding values are 12.75 for Hybrid, 14.95 for Rack+ResTree, 18.14 for Rack-only, and 21.87 for Pure Machine Learning (PureML). Additional drift diagnostics provide descriptive evidence of a temporal distribution shift over the studied interval, supporting the use of time-aware evaluation. A strictly causal offline replay also suggests that a conservative validation-gated model-refresh policy may improve future-window pointing accuracy under controlled conditions. Additional analyses of condition-wise robustness, feature relevance, and computational overhead support the practical feasibility of the framework.

关键词

Robotic telescopes/Telescope pointing/Astronomical instrumentation/Machine learning/Data analysis

Key words

Robotic telescopes/Telescope pointing/Astronomical instrumentation/Machine learning/Data analysis

引用本文复制引用

Weiqing Hu,Qingsheng Zhu.Hybrid rack-machine-learning pointing correction for a 1 m robotic telescope[EB/OL].(2026-11-18)[2026-11-19].https://chinaxiv.org/abs/202611.00187.

学科分类

天文学
首发时间: 2026-11-18
下载量:0
|
点击量:2
段落导航相关论文