Assembly constraint solver based on point-cluster reduction

Abstract

In recent years constraint-based modeling has become an important topic in the new generation CAD system. In this paper, an algorithm based on point-cluster reduction for assembly constraint satisfaction problem is presented. The algorithm combines graph-based with rule-based approaches and solves the assembly constraint satisfaction problem efficiently. Firstly, The simplified model of geometric constraint system is established and the graph representation of this model is proposed. Then the geometric reasoning algorithm based on point-cluster reduction is discussed in detail and the global constraint algorithm is presented based on the reduction algorithm Finally, the prototype of an assembly constraint solver is implemented using the object-oriented method. It is designed to be an important subsystem of a professional CAD system.