Research on the Tracking Registration Method of Markerless Augmented Reality for Product Assembly

Abstract

To solve the problems of assembly object variety, volume difference, and texture features not being obvious inthe intelligent products assembly process, augmented-reality technology was employed. To improve thefeasibility and robustness of the augmented-reality tracking registration algorithm in the product assemblyscenarios, combined with the weak texture of product assembly operation objects and the complex scenes, areal-time and robust unmarked tracking registration method was proposed. First, the multimodal approach ofLINEMOD was improved to detect the weakly textured assembly target base so that the rough pose of thecamera can be obtained and the problem of reinitialization can be solved when the process of trackingregistration is interrupted. Then, the iterative closest point (ICP) registration algorithm was improved toaccelerate the registration process while accurately positioning the camera pose to determine the benchmarkfor virtual-real fusion. Finally, based on the pose relationship between the assembly and the assembly guidanceinformation established in the offline phase, visual guidance of the assembly process was realized. Aperformance analysis and comparison of the proposed tracking registration method show that the proposedmethod has better accuracy than the LINEMOD template matching method. The visual display of the assemblyguidance process shows that the proposed method can be applied to the assembly scene of products lackingtexture to realize intelligent assembly.