Liaison-Based Enriched CAD Model Representation for Assembly Tasks

In mechanical engineering, the analysis and evaluation of CAD models of the final assembly are largely exploited to address complex assembly tasks to enhance the production process. However, CAD models definitely include geometric and topological information on parts, while the availability of additional important information, such as the semantics of parts, is not given for granted and is manually provided. This work aims at defining an enriched CAD model representation to integrate and provide product and assembly process information in a unique model to be exploited as input in assembly tasks. The representation is based on a new data structure, denoted as liaison, which is meant to comprehensively express the relation between two mating parts of the assembly. First, the key elements characterizing a liaison are pointed out, which are both geometric and semantic data regarding the assembly's parts and their relations. It is underlined that all the information stored in the proposed liaison-based representation is the outcome of a process that automatically extracts data from the product boundary representation as created/exchanged by commercial CAD systems. Then, the the paper describes the process to define and compute the list of liaisons, and thus returning the enriched CAD model representation. Finally, the developed interface for inspecting the automatically created enriched CAD model representation is discussed as well as possible applications to strengthen the potential of liaisons in optimizing assembly tasks.