面向智能测绘的测量平差课程改革:理念、体系与路径设计OA

In the era of intelligent geomatics,multi-source,massive,and high-dimensional observational data pose severe challenges to the traditional adjustment of measurements course.The current curriculum excessively emphasizes mathemati-cal derivations and idealized models,resulting in a disconnection from the dynamic and complex data processing tasks that students will encounter in practice.To this end,addressing the new requirements of intelligent geomatics,this paper restruc-tured the adjustment of measurements course systematically from two dimensions:content framework and implementation pathways,taking the"data science core"as the reform concept.In terms of the content framework,a three-tier progressive knowledge structure of"classical adjustment,modern data processing,and fundamentals of intelligent algorithms"was con-structed,and filtering and prediction,variance component estimation,and the application of machine learning in outlier detection were introduced.Regarding implementation pathways,an integrated"theory,code,and visualization"teaching model based on Python programming was promoted.Through real-world project cases such as the Global Navigation Satel-lite System(GNSS)network and inertial measurement unit(IMU)/digital elevation model(DEM)fusion,the data-driven modeling capabilities of students were cultivated.Simultaneously,a cloud-native computational experiment platform was developed to overcome the teaching bottleneck posed by computationally intensive algorithms.This reform aims to transform the adjustment of measurements from a tool-based course focusing on"mathematical optimal solutions"into a core course that cultivates the scientific thinking of"intelligent processing of spatial data,"thereby providing support for the training of new engineering talents in geomatics.