Title: Ai-Driven Robotic-Tool Selection for Draping Composite Preforms Based on a Geometric Surface Segmentation Approach.

Authors: Moritz Lennartz, Patrick Liebe, Hannah Dammers, Thomas Gries

DOI: 10.33599/nasampe/s.24.0113

Abstract: Existing automation solutions for the production of composite components are primarily designed for mass production, given their high investment costs and limited flexibility regarding the potential range of variants. As a result, a significant portion of all manufactured composite components is still produced manually. Due to a rapidly growing shortage of skilled workers across the industry, especially in small and medium-sized enterprises (SMEs), there is a strong demand for a flexible automation solution capable of accommodating a wide range of variants with minimal investment and setup costs. However, the flexible usage of robotics in composite industry SMEs is currently hindered by a lack of expertise in robot programming and AI. Therefore, we introduce an approach for flexible automation in robot-based draping using AI. The aim is to enable a robot to capture the geometric properties of any mold so that a suitable draping tool can be selected and an automated draping step performed. Starting with a point cloud-based capture of the mold, a supervised learning model is developed to categorize the mold surface into individual geometry classes. Afterwards, a model can be created to select a suitable draping tool for the robot for each of the identified classes. This marks an important step towards enabling fully automated draping.

References: [1] A. Lenne, JEC Observer – Overview of the global composites market, 2021 – 2026. France, 2022 [2] H. Krieger, Methode zur Auslegung von Gelegen mit local angepassten Fertigungsparametern für Hochleistungs-Faserverbundkunststoffe. Düren: Shaker, 2015 [3] S. Schöfer, Kraftgesteuerte Materialzuführung beim Drapierprozess in der automatisierten Herstellung von Faserverbundwerkstoffen. Düren, Shaker, 2020 [4] M. Christ, Definition und Quantifizierung der Drapierbarkeit von multiaxialen Gelegen durch die Vermessung von Einzeleffekten. Bremen, 2018 [5] F. Reux, M. Mikdam, Overview of the Global Composite Market 2018 – 2023. France, 2019 [6] M. Elkington, D. Bloom, C. Ward, A. Chatzimichali, K. Potter, Hand layup: understanding the manual process. Advanced Manufacturing: Polymer & Composites Science, vol. 1, no. 3, pp. 138–151, 2015, DOI: 10.1080/20550340.2015.1114801 [7] H. Dammers, M. Behery, M. van Gemmeren, B. Manin, G. Lakemeyer, T. Gries, Supervised Learning for Robotic Draping Tasks in Composite Preforming. In: Proceedings of the International Conference on Composite Materials (ICCM), Belfast, 2023 [8] C. Eitzinger, C. Frommel, S. Ghidoni, E. Villagrossi, System concept for human-robot collaborative draping. Baden/Zürich, Schweitz, 2021. [9] S. Asenkerschbaumer, H. Kagermann, T. Klüwer, K. O. Arras, R. Hartke, A. Kunack, F. Süssenguth, Innovationspotenziale KI-basierter Robotik. München, 2023, DOI: https://doi.org/10.48669/aca_2023-15 [10] R. K. Malhan, A. M. Kabir, A. V. Shembekar, B. Shah, S. K. Gupta, T. Centea, Hybrid Cells for Multi-Layer Prepreg Composite Sheet Layup. In: 2018 IEEE 14th International Conference on Automation Science and Engineering (CASE), München, 2018, DOI: https://doi.org/10.1109/COASE.2018.8560586 [11] R. K. Malhan, A. V. Shembekar, A. M. Kabir, P. M. Bhatt, B. Shah, S. Zanio, S Nutt, S. K. Gupta, Automated planning for robotic layup of composite prepreg, in: Robotics and Computer-Integrated Manufacturing, vol. 67, 2021, DOI: https://doi.org/10.1016/j.rcim.2020.102020 [12] S. Minaee, Y. Boykov, F. Porikli, A. Plaza, N. Kehtaranavaz, D. Terzoploulos, Image Segmentation Using Deep Learning: A Survey. In: IEEE transactions on pattern analysis and machine intelligence, vol. 7, P. 3523–3542, 2022, DOI: 10.1109/TPAMI.2021.3059968 [13] X. Liu, Z. Deng, Y. Yang, Recent progress in semantic image segmentation. In: Artificial Intelligence Review, vol. 2, P. 1089-1106, 2019, DOI: https://doi.org/10.1007/s10462-018-9641-3 [14] Y. Guo, H. Wang, Q. Hu, H. Liu, L. Liu, M. Bennamoun, Deep Learning for 3D Point Clouds: A Survey, 2019, Available: https://arxiv.org/abs/1912.12033 [15] C. R. Qi, H. Su, K. Mo, L. Guibas, PointNet : Deep Learning on Point Sets for 3D Classification and Segmentation, 2016: Available: https://arxiv.org/abs/1612.00593 [16] C. R. Qi, L, Yi, H. Su, L. Guibas, PointNet++ : Deep Hierarchical Feature Learning on Point Sets in Metric Space, 2017, Available: https://arxiv.org/abs/1706.02413 [17] P. S. Wang, OctFormer: Octree-based Transformer for 3D Point Clouds, 2023, Available: https://arxiv.org/abs/2305.03045 [18] A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, I. Polosukhin, Attention is all you Need. 2017, Available: https://arxiv.org/abs/1706.03762 [19] J. G. Lambourne, K. D. D. Willis, P. K. Jayaraman, A. Sanghi, P. Meltzer, H. Shayani, BRepNet: A Topological Message Passing System for Solid Models. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2021, P. 12773-12782, Available: https://arxiv.org/abs/2104.00706 [20] F. Zhuang, Z. QI, K. Duan, D. Xi, Y. Zhu, H. Zhu, H. Xiong, Q. He, A Comprehensive Survey on Transfer Learning, 2019, Available: https://arxiv.org/abs/1911.02685

Conference: SAMPE 2024

Publication Date: 2024/05/20

SKU: TP24-0000000113

Pages: 13

Price: $26.00