Path Planning of a Type of Porous Structures for Additive Manufacturing

Porous structures are widely used in our daily life due to their special properties such as higher rigidity and lightweight. With the popularity of additive manufacturing (AM), producing porous structures by AM shows great advantages over traditional manufacturing technologies. In this paper, we first introduce a framework to model a type of porous structures which contain lots of small holes. Then we propose an efficient path planning algorithm for this type of porous structures. The algorithm consists of the following steps: for each slice of a porous structure, we first subdivide the domain of the slice into subdomains by generalized Voronoi diagram such that each subdomain contains exactly one hole, and then we carry out dual operation to obtain a partition of the domain into a set of subregions. A route is then found to traverse every subregion by solving a traveling salesman problem (TSP) using genetic algorithm (GA), and along the route subregions are merged into larger ones to improve printing efficiency. Finally we fill each merged subregion with a single Fermat spiral curve and further optimize smoothness and uniformity of the filling path. We demonstrated a variety of testing examples and the experimental results showed the superiority and effectiveness of our method in terms of material cost, printing time and structure stability.