ABSTRACT Augmented reality (AR) technology enables the interaction between physical objects and virtual objects, which enriches the expression forms of virtual information. At present, most relevant research focuses on the projection and positioning of AR applications. However, the interaction precision of AR applications is generally low due to the limited computing capability of related devices. Based on this, this paper proposes an octree‐based dynamic voxel grid representation method. Virtual objects are constructed with voxel models, and during interaction, the voxel grids of the virtual model are first merged and then decomposed in real time. This method can reduce the substantial memory overhead caused by the improvement of model accuracy, improve computational efficiency in the high‐precision interaction process, and ensure the real‐time performance of interaction. By integrating this method into the field of mechanical machining, we have constructed a hybrid robotic machining trajectory verification system. Experimental results demonstrate that this method can improve computational efficiency and realize high‐precision interaction.
Wang et al. (2026) studied this question.