人工智能大地测量现状与趋势研究

Current Status and Trends of Artificial Intelligence Geodesy

  • 摘要: 人工智能(artificial intelligence,AI)作为新一轮科技革命的核心驱动力,正在深刻重塑大地测量学的研究范式与技术体系。在深入总结大地测量学科的重要进展和关键应用领域基础上,重点分析了深度学习与机器学习等AI算法在坐标框架维持、高程与深度基准构建、重力场建模、导航定位以及地质灾害形变监测等核心领域的最新动态,梳理形成了人工智能大地测量(artificialintelligence geodesy,AIG)技术新方向及其发展趋势。AIG将突破传统大地测量方法的局限性,利用累积的大地测量观测数据集、样本数据集、产品数据集,构建海量多源异构测量数据智能处理算法,形成复杂测量误差改正和多重物理场计算的AIG大模型,驱动大地测量学科从“大地测量数据驱动”的优化估计向“物理机理与数据智能双驱动”建模决策转型,支撑国家空间基准维持和全域高精度智能定位导航授时等服务,满足自然资源治理、低空经济、防灾减灾等应用需求。

     

    Abstract: Objectives: Artificial Intelligence (AI), acting as the core driving force of the current technological revolution, is profoundly reshaping the research paradigms and technical systems of Geodesy. The primary objective of this research is to systematically define the concept of Artificial Intelligence Geodesy (AIG) and analyze its developmental trends. It aims to clarify how AI technologies can upgrade traditional geodetic frameworks to facilitate the transformation from data processing to intelligent decision-making, thereby meeting emerging strategic demands in natural resource governance and the low-altitude economy. Methods: A systematic review is conducted to investigate the application status and latest progress of Deep Learning and Machine Learning algorithms within the geodetic domain. The research focuses on analyzing five core areas: coordinate frame maintenance, height and depth datum construction, gravity field modeling, navigation and positioning and geological disaster deformation monitoring. By utilizing accumulated geodetic observation datasets, sample datasets, and product datasets, the technical architecture of AIG is examined. The methodology involves constructing intelligent processing algorithms for massive multi-source heterogeneous survey data to address complex geodetic challenges. Results: The investigation reveals that AIG technology significantly breaks through the limitations of traditional geodetic methods. By integrating advanced AI algorithms, AIG large models are formed, which are capable of performing complex measurement error correction and multiphysics field calculations. A critical finding is that the integration of AI drives a fundamental transformation of the Geodesy discipline from a "geodetic data-driven" optimization estimation approach to a "dual-driven paradigm combining physical mechanisms with data intelligence." These models demonstrate superior capability in handling non-linear problems and massive heterogeneous data compared to conventional approaches. Conclusions: AIG represents a pivotal new direction for the future of Geodesy. It provides robust technical support for essential services such as national spatial reference maintenance and alldomain high-precision intelligent Positioning, Navigation, and Timing (PNT). Furthermore, the advancement of AIG is essential for satisfying the rigorous application demands of disaster prevention and reduction, as well as the rapidly growing low-altitude economy. Future developments will focus on deepening the dual-driven modeling approach to further enhance the precision, reliability, and intelligence of geodetic services.

     

/

返回文章
返回