Deep learning enabled invisibility cloak design

• Main Author: Yang, Chenbo
• Other Authors: Luo Yu
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2022
• Subjects: Engineering::Electrical and electronic engineering::Optics, optoelectronics, photonics
• Online Access: https://hdl.handle.net/10356/158016
• Institution: Nanyang Technological University

The invisibility cloak can be seen in numerous films and television shows; it is a fictitious device that can conceal the item inside it. The invisibility cloak has been a popular study issue in many sectors, particularly in the military such as stealth fighters. The transformation optics enables people to design new materials by using the coordinate transformation method to control the way of light propagation, thereby realizing invisibility. However, invisibility cloak based on this method performs poorly in describing the performance of the multilayered cloak in practical situations. In this project, the performance of a multilayered invisibility cloak will be modeled by Matlab and some deep learning based existing Matlab optimization tools will be used to optimize the model, in order to dim the intrinsic scatterings caused by discretization and simplification. In most instances, the optimized model not only has better performance, but also only requires a few layers, which is easy to implement in actual manufacturing.