Robotized Additive Manufacturing for Silicone assisted by an Artificial Intelligence – RAMSAI
Robotized Additive Manufacturing for Silicone assisted by an Artificial Intelligence
/
Challenges and objectives
The factory of the future must ensure flexible, reliable, and efficient production. For the production of silicone parts, current processes based on molding and injection techniques do not meet these requirements. For this flexible material (silicone), the rheological behavior during deposition in Additive Manufacturing (AM) is complex and difficult to model. Conventional AM strategies with offline adjusted deposition based on a priori knowledge models then find their limits. In the RAMSAI project, the team proposes to develop a new approach to AM by exploiting a multidisciplinary approach, combining materials science, mechatronics, artificial intelligence, and robotics. The project is structured around three axes: 1) Characterize and model the rheological behavior of silicone during deposition using an AI-based approach. 2) Supervise and control manufacturing to increase productivity and process robustness. A closed-loop approach integrating a silicone behavior model will correct small printing errors. 3) Optimize silicone AM for complex parts using a robotic arm and a variable geometry nozzle. The targeted application area will be healthcare, with a proven need for tools for producing realistic anatomical models that are patient-specific for training and planning surgical interventions.
/
The significant results at this stage of the project mainly concern: 1) The characterization of the rheological behavior of silicone, and the determination of the silicones best suited to 3D printing (defined by a set of parameters that will determine the ability of the silicone to be used in additive manufacturing). 2) The development of an experimental additive manufacturing environment integrating a serial robotic arm equipped with a silicone print head. 3) Software development allowing, from an STL file, to obtain the robotic trajectories for the layer-by-layer deposition of silicone. 4) The development of metrics to carry out the supervision of 3D printing. Currently, several indicators are being considered, including the error between the actual height of the part and the theoretical height using a profilometer. The scientific and technological developments carried out will also lead to the creation of demonstrators that will highlight the unprecedented performance of the process.
/
Mosser, L.; Barbé, L.; Rubbert, L.; Renaud, P. ROS2 for soft materials additive manufacturing. ROSConFr. Bordeaux, France. June 2023.
Sand, J.; Wach, B.; Bednarczyk, M.; Barbé, L.; Geiskopf, F. Robotized Additive Manufacturing of Silicone for Skeleton-Reinforced Linear Soft Actuators. 6th IEEE-RAS International Conference on Soft Robotics. Singapore. April 2023.
The factory of the future must ensure flexible, reliable and efficient production. For the production of silicone parts, widely used in industry, current processes based on molding and injection techniques do not meet these requirements. Additive manufacturing (AM) should overcome these problems. However, the rheological behavior of silicone as a flexible material during deposition is complex and difficult to model. Conventional additive manufacturing strategies, using deposition strategies which are adjusted off-line with a priori knowledge models, find their limits. In the RAMSAI project, we propose to develop a new approach to AM by exploiting artificial intelligence to control the rheological behavior of silicone during extrusion. The work will be multidisciplinary, combining material science, mechatronics, artificial intelligence, and robotics. A new silicone printing head integrating rheological control of the silicone and dimensional control of the filament will be developed. The closed-loop control of silicone rheological behavior will use physics-based machine learning and predictive algorithms. The filament dimensions will be controlled using a variable shape nozzle already developed by the project partners. For complex parts, the contribution of a control of the print head in position and orientation using a robotic arm will be determined. These scientific and technological developments will lead to the realization of demonstrators which will highlight the new performances of the process. The first application area targeted will be health care, with a proven need today for tools to produce patient-specific realistic anatomical models for training and planning surgical interventions.
Project coordination
Laurent Barbé (Laboratoire des sciences de l'ingénieur, de l'informatique et de l'imagerie)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
Partnership
ENSAM - PIMM Ecole Nationale Supérieure d'Arts et Métiers - Procédés et Ingénierie en Mécanique et Matériaux
ICBMS Institut de Chimie et de Biochimie Moléculaires et Supramoléculaires
IMP Ingénierie des Matériaux Polymères
ICube Laboratoire des sciences de l'ingénieur, de l'informatique et de l'imagerie
Help of the ANR 639,777 euros
Beginning and duration of the scientific project:
January 2023
- 48 Months