It is crucial to investigate the urban agglomerations spatio-temporal evolution patterns and driving factors for analyzing the urban spatial structure-functional division and promoting the coordinated development of urban agglomerations. In this study, a novel vegetation-building-nighttime light-adjusted index (VBNAI) was established for rapid and effective mapping of urban construction land (UCL) in Central Plains Urban Agglomeration (CPUA), China during 2000–2020 based on Google Earth Engine (GEE) platform. Compared with traditional indices, VBNAI can significantly decrease the blooming effect, Normalized Difference Vegetation Index (NDVI) saturation, and soil background of nighttime light data. In addition, the urban expansion indices and standard deviation ellipse model were synthetically adopted to analyze the spatio-temporal evolution pattern of urban expansion. The gravity model and the geographically weighted regression model were employed to determine the spatial interaction forces and drivers of urban expansion, respectively. The results showed that the VBNAI index has obvious advantages in efficiency and accuracy to extract UCL with the overall accuracy of more than 91%. The UCL of CPUA had increased by 4489.84 km² during 2000–2020 with the gravity center moving towards southeast continuously. From 2000 to 2010, the urban expansion was in a ‘center-hinterland’ pattern which had benefit from the favorable effect of the traffic shaft belt. During 2010–2020, the urban network structure had basically established. Urban expansion had been influenced by a variety of socio-economic and demographic factors, and the impact degree varied from region to region. This study could provide scientific references for facilitating the intensive utilization of urban resources and optimizing the spatial development pattern of urban agglomeration.