3D CAD model retrieval based on GA-ACO

To reuse 3D CAD models more efficiently, a 3D CAD model retrieval approach which combines Genetic Algorithm (GA) and Ant Colony Optimization (ACO) is proposed. A CAD model is represented by feature dependency graph (FDG). The similar sub-graphs in the corresponding FDGs of different models are obtained according to polymorphic combinatorial GA and ACO. GA is adopted to obtain suboptimal solutions. The pheromone of ACO is initialized according to the suboptimal solutions, and then a further search among the suboptimal solutions is operated for the better solution. And finally, the optimal solutions of the product design can be searched. The hybrid approach is accomplished in convergence efficiency and solution precision. Our experimental results show that, the algorithm can seek the modeling containing detail characteristics and satisfy user´s personalized need.