Robot-Assisted Additive Manufacturing

Additive Manufacturing (AM) has a central role in actual industry that aims at profitability, efficiency, performance and customization by merging the digital and physical world. AM gives the possibility to fabricate unique or small batch size parts on-demand, without outsourcing and with competitive lead times by converting the digital model of the geometry directly in the corresponding physical object. Meaning thatAMis enabling the possibility of prototype components and mechanisms reducing the lead time needed to outsource the production and to reduce the overall costs. AM technologies are capable to manage more materials, from polymers and ceramics to metals and composites with different feedstocks. Although these technologies are different in the process, they have in common that the fabrications happens layer-by-layer. The "traditional" layer-bylayer fabrication consists in the superimposition of 2D-slices with a predefined thickness sticked together to realize the final component. This way of proceed has been borrowed by the initial development of CAM software used for subtractive operations using 3-axis machines or 3 degrees of freedom machines (DoFs). However, the natural evolution of subtractive processes brought the use of machines with even more than 5-axis to process geometries with complex surfaces and curves. Employing Industrial robots in AM enable material deposition along the 3D space enabling the possibility to overcome the limits of traditional slicing techniques. In recent years, researchers have been investigating the use of more DoFs to improve AM process enabling supportless fabrication for overhanging geometry and to improve the surface quality. When facing overhanged geometry, the deposition layer-by-layer is not effective because the layer starts to not be self supported but suspending in the air. This issue introduces an increased anisotropy and besides, in most of the cases leads to the failure of the fabrication. Traditional slicing usually introduces support structures with the aim that every portion of every layer is supported by the previous one. However, this introduces an increased time for fabrication and an increase of the material consumption which is used to generate the supports. Novel approaches called non-planar slicing aim to solve the current problem using more DoFs such as using industrial robots, to deposit material in the entire space and not just in 2D. This thesis inquires novel approaches in the use of AM developing new slicing techniques and exploiting industrial robots for the realization of the components. Anthropomorphic robots which are robotic systems with 6 DoFs allows to define the end effector pose in the complete 3D space (position and orientation of the coordinate system associated with this body) as well as offer larger workspace compared to the machine size. This implies the need for software able to completely exploit the possibilities of these systems by generating suitable toolpaths according to the process itself (reference substrate, layer, initial CAD model and parameters) and the specific cell at hand (tool collision and singularities). The research activity focused on the explore on how to integrate robotic systems in AM process, starting from the hardware integration and the programming strategies for following trajectories and then the main topic which is focusing on developing different non-planar slicing technique aim to minimize the layer thickness variation in free form geometries. All the methods shown use simulation to check the feasibility of the trajectory planning and then for most of the methods the experimental activities for the fabrication has been carried out.