Design of Modular Origami Structures for Multifunctional Cloaking and Protection

Project begins 2022.

The bodies of ground combat vehicles are increasingly required to serve various roles including structural load bearing, blast mitigation, thermal management, acoustic cloaking, visual camouflage and more. In recent years, the principles of origami have enabled different engineering systems that can deploy, reconfigure, and adapt for multiple functions. For example, the PI has demonstrated stiff and lightweight deployable structures, improved energy absorption, and adaptive thermal performance, while other work has shown acoustic cloaking, adaptive electro-magnetic responses, and tunable transparency. While many of the underlying geometries of these advanced origami systems are similar, it remains challenging to identify, pick, and design specific origami systems to fulfil multiple performative functions.

The overarching goal of this research is to harness novel advancements and adaptability of origami structures to achieve multiple functions for cloaking and protection. Our work will first need to establish efficient simulation methods, and then use them for generative ML design of the multifunctional origami. We will create and combine numerical models into a single framework that can efficiently capture acoustic, thermal, and mechanical behaviors of the origami. The models will holistically evaluate and quantify the multi-physical behaviors of the origami throughout the deployment path. These new analytical tools will allow us to explore the fundamental behavior of reconfigurable origami in the physical world where large deformations, mechanics, and multi-physics are often coupled (e.g. thermal boundaries change depending on the 3D configuration).

Next, we will need to create a framework and appropriate methods for ML design of origami. We will establish a database to characterize the influence of origami design parameters on the multi-objective and adaptive performance. We will then adapt modern ML methods to exploit the unlimited design space offered by different origami patterns and variations. In particular, the methods will systematically compare the quantitative and qualitative performance of origami designs that are categorically different (meaning two origami may share almost no common features in geometry or intrinsic design). The final designs will achieve multiple functions of cloaking and protection. The proposed database and ML design approach can revolutionize all future origami design, which is currently complicated by the categorically different systems, large geometric 3D reconfigurations, and multiple design objectives.

Prior publications from related work:

• Wo, Z., and Filipov, E.T. (2023) “Stiffening multi-stable origami tubes by outward popping of creases,” Extreme Mechanics Letters, Vol. 58. 101941
• Zhu, Y. and Filipov, E.T. (2022) “Harnessing Interpretable Machine Learning for Origami Feature Design and Pattern Selection,” Scientific Reports, Vol. 12, No. 19277

Publications:

• Zhu, Y., Filipov, E.T. Large-scale modular and uniformly thick origami-inspired adaptable and load-carrying structures. Nat Commun 15, 2353 (2024). https://doi.org/10.1038/s41467-024-46667-0

#3.21