Contact state estimation based on surface-matching in virtual assembly

Abstract

Contact problems are one of the most challenging fields in virtual assembly. Information of contact states could be utilized to realize compliant motion of work pieces, to analyze the contact stress, to assist positioning parts and so on. Some methods have already been proposed to estimate contact states between objects and in most of these methods contact states between objects are simplified in order to realize real-time visual reality animation. While in virtual assembly contact states between parts are required to analyze contact stress, deformation and quality. Besides the contact state estimation method for virtual assembly should be able to handle a number of complex parts in real time. There are rarely known methods which could meet this requirement till now. In this study a contact state estimation algorithm based on surface-matching for virtual assembly is proposed. Contacts between parts are categorized into six basic types according to contact region of surfaces. Based on continuous collision detection of polyhedral models a novel contact state identification algorithm which is based on surface matching is proposed. Then contact evolution algorithm, which utilizes the extern force and contact information, is implemented to handle evolution of contact state. Finally a prototype system is developed to verify the above technologies. Experiment results reveal that contact state between parts could be estimated correctly in real time virtual assembly. The proposed contact state estimation algorithm provides a complete solution to obtain the contact state between parts in virtual assembly. Information of contact state between parts could be utilized to realize contact dynamic, contact stress analysis, assembly quality analysis, and so on.