Reinforcement learning for block decomposition of planar CAD models
Engineering with Computers - Trang 1-11 - 2024
Tóm tắt
The problem of hexahedral mesh generation of general CAD models has vexed researchers for over 3 decades and analysts often spend more than 50% of the design-analysis cycle time decomposing complex models into simpler blocks meshable by existing techniques. The decomposed blocks are required for generating good quality meshes (tilings of quadrilaterals or hexahedra) suitable for numerical simulations of physical systems governed by conservation laws. We present a novel AI-assisted method for decomposing (segmenting) planar CAD (computer-aided design) models into well shaped rectangular blocks. Even though the simple examples presented here can also be meshed using many conventional methods, we believe this work is proof-of-principle of a AI-based decomposition method that can eventually be generalized to complex 2D and 3D CAD models. Our method uses reinforcement learning to train an agent to perform a series of optimal cuts on the CAD model that result in a good quality block decomposition. We show that the agent quickly learns an effective strategy for picking the location and direction of the cuts and maximizing its rewards. This paper is the first successful demonstration of an agent autonomously learning how to perform this block decomposition task effectively, thereby holding the promise of a viable method to automate this challenging process for more complex cases.
Tài liệu tham khảo
Wang E, Nelson T, Rauch R (2004) Back to elements - tetrahedra vs hexahedra. In: Proceedings of the 2004 International Ansys Conference. https://www.ansys.com/-/media/ansys/corporate/resourcelibrary/conference-paper/2004-int-ansys-conf-9.pdf
Pietroni N, Campen M, Sheffer A, Cherchi G, Bommes D, Gao X, Scateni R, Ledoux F, Remacle J-F, Livesu M (2022) Hex-mesh generation and processing: a survey. ACM Transactions on Graphics 42(2):1–44
Sarrate, J., Ruiz-Gironés, E., Roca, X.: Unstructured and semi-structured hexahedral mesh generation methods. Computational Technology Reviews 10 (2017)
Shephard MS, Georges MK (1991) Automatic three-dimensional mesh generation by the finite octree technique. International Journal for Numerical Methods in Engineering 32(4):709–749
Owen S (2016) An Introduction to Automatic Mesh Generation Algorithms. Short Course Notes International Meshing Roundtable, Washington, D.C. . https://www.osti.gov/servlets/purl/1394107
Thompson JF (1987) A general three-dimensional elliptic grid generation system on a composite block structure. Computer Methods in Applied Mechanics and Engineering 64:377–411
William G, Hall C (1973) Construction of curvilinear coordinate systems and their application to mesh generation. International Journal of Numerical Methods in Engineering 7(4):461–477
White, D.R., Mingwu, L., Benzley, S.E., Sjaardema, G.D.: Automated hexahedral mesh generation by virtual decomposition. In: Proceedings of the 4th International Meshing Roundtable, pp. 165–176 (1995)
White DR, Saigal S, Owen SJ (2004) Ccsweep: automatic decomposition of multi-sweep volumes. Engineering with Computers 20:222–236
Wang R, Shen C, Chen J, Wu H, Gao S (2017) Sheet operation based block decomposition of solid models for hex meshing. Computer-Aided Design 85:123–137
Gordon WJ, Thiel LC (1982) Transfinite mappings and their application to grid generation. Applied Mathematics and Computation 10–11:171–233
Mingwu, L., Benzley, S.E., Sjaardema, G., Tautges, T.: A multiple source and target sweeping method for generating all hexahedral finite element meshes. In: Proceedings of 5th International Meshing Roundtable, pp. 165–176 (1996)
Shepherd, J.F., Mitchell, S.A., Knupp, P.M., White, D.R.: Methods for multisweep automation. In: Proceedings of the 9th International Meshing Roundtable, pp. 77–87 (2000)
Lu, Y., Gadh, R., Tautges, T.J.: Volume decomposition and feature recognition for hexahedral mesh generation. In: 8th International Meshing Roundtable, Lake Tahoe, CA, pp. 269–280 (1999)
White DR, Tautges TJ (2000) Automatic scheme selection for toolkit hex meshing. International Journal of Methods in Engineering 49(1):127–144
Lu JH-C, Quadros WR, Shimada K (2017) Evaluation of user-guided semi-automatic decomposition tool for hexahedral mesh generation. Journal of Computational Design and Engineering 4(4):330–338
Blum, H.: A transformation for extracting new descriptors of shape. In: Models for the Perception of Speech and Visual Forms, pp. 362–380 (1967)
Price MA, Armstrong CG, Sabin MA (1995) Hexahedral mesh generation by medial surface subdivision: part i. solids with convex edges. International Journal of Numerical Methods in Engineering 38(19):3335–3359
Zhang, Z., Wang, Y., Jimack, P.K., Wang, H.: Meshingnet: A new mesh generation method based on deep learning. Computational Science – ICCS 2020. Lecture Notes in Computer Science 12139 (2020)
Huang, K., Krügener, M., Brown, A., Menhorn, F., Bungartz, H.-J., Hartmann, D.: Machine Learning-based optimal mesh generation in computational fluid dynamics. arXiv preprint arxiv:2102.12923 (2021)
Dielen, A., Lim, I., Lyon, M., Kobbelt, L.: Computing direction fields for quad mesh generation. In: Eurographics Symposium on Geometry Processing, vol. 40 (2021)
Pak, D., Liu, M., Kim, T., Liang, L., McKay, R., Sun, W., Duncan, J.: Distortion energy for deep learning-based volumetric finite element mesh generation for aortic valves. Medical Image Computing and Computer Assisted Intervention – MICCAI 2021. Lecture Notes in Computer Science 12906, 485–494 (2021)
Bohn J, Feischl M (2021) Recurrent neural networks as optimal mesh refinement strategies. Computers & Mathematics with Applications 97:61–76
Yang, J., Dzanic, T., Petersen, B., Kudo, J., Mittal, K., Tomov, V., Camier, J.-S., Zhao, T., Zha, H., Kolev, T., et al.: Reinforcement learning for adaptive mesh refinement. In: International Conference on Artificial Intelligence and Statistics, pp. 5997–6014 (2023)
Wu T, Liu X, An W, Huang Z, Lyu H (2022) A mesh optimization method using machine learning technique and variational mesh adaptation. Chinese Journal of Aeronautics 35(3):27–41
Pan J, Huang J, Wang Y, Cheng G, Zeng Y (2021) A self-learning finite element extraction system based on reinforcement learning. Artificial Intelligence for Engineering Design, Analysis and Manufacturing 35:180–208
Pan, J., Huang, J., Cheng, G., Zeng, Y.: Reinforcement learning for automatic quadrilateral mesh generation: a soft actor-critic approach. arXiv report (2022). https://arxiv.org/abs/2203.11203
Tong H, Qian K, Halilaj E, Zhang YZ (2023) SRL-assisted AFM: Generating planar unstructured quadrilateral meshes with supervised and reinforcement learning-assisted advancing front method 72:102109
Manevitz LM, Yousef M, Givoli D (2002) Finite-element mesh generation using self-organizing neural networks. Computer-Aided Civil and Infrastructure Engineering 4(12):233–250
Çinar, A., Arslan, A.: Neural networks based mesh generation method in 2-d. Lecture Notes in Computer Science 2510 (EurAsia-ICT 2002: Information and Communication Technology), 395–401 (2002)
Kim YS (1992) Recognition of form features using convex decomposition. Computer-Aided Design 24(9):461–476
Wu H, Gao S, Wang R, Chen J (2018) Fuzzy clustering based pseudo-swept volume decomposition for hexahedral meshing. Computer-Aided Design 96:42–58
Takata O et al (1999) A knowledge-based mesh generation system for forging simulation. Applied Intelligence 11(2):149–168
Qin, F., Li, L., Gao, S., et. al.: A deep learning approach to the classification of 3d cad models. Journal of Zhejiang Univ. - Science C 15, 91–106 (2014)
Bronstein, M.M., et.al.: Geometric deep learning: Going beyond euclidean data. IEEE Signal Processing Magazine 34(4) (2017)
Boussuge, F., Tierney, C.M., Robinson, T.T., Armstrong, C.G.: Application of tensor factorisation to analyse similarities in cad assembly models. In: Proceedings of the 28th International Meshing Roundtable (2019)
Wu, R., X., C., , Zheng, C.: DeepCAD: A deep generative network for computer-aided design models. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (2021)
Lin, C., Fan, T., Wang, W., Nießner, M.: Modeling 3d shapes by reinforcement learning. In: Computer Vision – ECCV 2020. Lecture Notes in Computer Science(), vol. 12355 (2020). https://doi.org/10.1007/978-3-030-58607-2_32
Danglade F, Pernot J-P, Véron P (2014) On the use of machine learning to defeature cad models for simulation. Computer-Aided Design and Applications 11(3):358–368
Owen, S., Shead, T.M., Martin, S.: CAD defeaturing using machine learning. In: Proceedings of the International Meshing Roundtable (2019)
Sutton RS, Barto AG (2018) Reinforcement Learning: An Introduction. A Bradford Book, Cambridge, MA, USA
Mnih V, Kavukcuoglu K, Silver D, Rusu AA, Veness J, Bellemare MG, Graves A, Riedmiller M, Fidjeland AK, Ostrovski G, Petersen S, Beattie C, Sadik A, Antonoglou I, King H, Kumaran D, Wierstra D, Legg S, Hassabis D (2015) Human-level control through deep reinforcement learning. Nature 518(7540):529–533. https://doi.org/10.1038/nature14236
Sahba, F., Tizhoosh, H.R., Salama, M.M.A.: A reinforcement learning framework for medical image segmentation. In: The 2006 IEEE International Joint Conference on Neural Network Proceedings, pp. 511–517 (2006). doi: 10.1109/IJCNN.2006.246725
Han, J., Yang, L., Zhang, D., Chang, X., Liang, X.: Reinforcement cutting-agent learning for video object segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2018)
Wang, Y., Dong, M., Shen, J., Wu, Y., Cheng, S., Pantic, M.: Dynamic face video segmentation via reinforcement learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2020)
Tiator, M., Geiger, C., Grimm, P.: Point cloud segmentation with deep reinforcement learning. In: 24th European Conference on Artificial Intelligence 2020, pp. 2768–2775 (2020)
Mukherjee N (2014) An art gallery approach to submap meshing. Procedia Engineering 82:313–324
OpenCascade: OpenCascade Technology, 7.7.0dev. OpenCascade.com, Moulineaux, France (2022). OpenCascade.com. https://dev.opencascade.org/doc/overview/html/
Haarnoja, T., Zhou, A., Abbeel, P., Levine, S.: Soft actor-critic: Off-policy maximum entropy deep reinforcement learning with a stochastic actor. CoRR abs/1801.01290 (2018) 1801.01290
Fey, M., Lenssen, J.E., Weichert, F., Müller, H.: SplineCNN: Fast geometric deep learning with continuous B-spline kernels. CoRR abs/1711.08920 (2017) 1711.08920
CUBIT: The Cubit Geometry and Mesh Generation Toolkit. Sandia National Laboratories, Albuquerque, NM, USA (2022). Sandia National Laboratories. https://cubit.sandia.gov/files/cubit/16.06/help_manual/WebHelp/cubithelp.htm
Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., Desmaison, A., Köpf, A., Yang, E., DeVito, Z., Raison, M., Tejani, A., Chilamkurthy, S., Steiner, B., Fang, L., Bai, J., Chintala, S.: Pytorch: An imperative style, high-performance deep learning library. In: Proceedings of the 33rd International Conference on Neural Information Processing Systems, pp. 8024–8035 (2019)
Fey, M., Lenssen, J.E.: Fast graph representation learning with PyTorch Geometric. In: ICLR Workshop on Representation Learning on Graphs and Manifolds (2019)