With the rapid development of positioning and 5G technology, mobile objects in real life provide timely updates of location information, leading to the generation of large amounts of mobile data. However, the vast majority of mobile objects often move within a fixed spatial area, with only a few mobile objects leaving this area. If the single time-parameterized R-tree (TPR-tree) index structure is used, it will update frequently due to these “escaped” moving objects, thereby affecting the query efficiency. To address this issue, a hybrid index structure is proposed based on a three-dimensional (3D)-grid TPR-tree. A 3D grid is used to quickly locate and prune query points or regions in 3D space, and a global TPR-tree is applied to reduce the impact of the moving object update frequency on the query efficiency. The experimental results showed that the hybrid index structure based on the 3D-grid TPR-tree yielded good performances in both mobile queries and continuous k-nearest neighbor queries.