In this study, a new square-based fiber Bragg grating (FBG) sensor network model is proposed to address possible link failures in FBG sensor networks and improve their reliability. Graph theory and optical switching are simultaneously applied to these sensor networks to improve their self-healing ability; the FBG sensor network is regarded as a directed graph. Three commonly used self-short-circuit algorithms are compared in terms of the self-healing capabilities that they provide to the optical fiber sensor network. Among these, the shortest-path faster algorithm achieved a high, nearly 90% repair accuracy and had an average repair time of 0.103 s, the shortest in this study. The newly designed FBG self-healing network can be reorganized and repaired when local damage occurs, thereby improving its reliability.

The optimized cloaking design for conducting cylinders of different sizes is studied based on the Mie scattering theory. We construct a concentric multi-layered cloak made of alternating materials with isotropic dielectrics and epsilon-near-zero (ENZ) material, the thickness of which can be determined through genetic algorithm. As the radius of the conducting cylinder increases, high order scattering contributions are becoming evident, and more layers are needed. The scattering cross sections of three different radii of PEC cylinders are minimized by utilizing different numbers of multi-layers respectively. We find that eight or less optimized layers can cancel most of the scattering from a conducting cylinder with its dimension compared to wavelength, and more effectively when taking the ENZ material as the inner starting shell. The frequency dependence of total scattering is also studied, leading to the result that the bandwidth decreases as the size of concealed PEC cylinder increases. Furthermore, it is shown that the cloaking efficiency is less sensitive to the permittivity and thickness of the ENZ material, due to the small phase variation in the ENZ material. The multi-layered cloak designed for a PEC target can also be used to evidently reduce the scattering of a dielectric core and design a multi-layered elliptical cloak.

This study proposes a long‐period fibre grating (LPFG) curvature estimation method based on random forest regression (RFR) to address the shortcomings of the existing curvature evaluation method, namely, polynomial fitting; these shortcomings cause difficulty in achieving adequate model regularity and application universality. The resonant wavelength and resonant peak amplitude of the LPFG are used as input variables in this method to develop an RFR model for curvature estimation, allowing for accurate curvature prediction of the sample. The results show that the RFR‐based LPFG curvature prediction model can better characterise the input–output regression relationship than back‐propagation neural networks. The average R ² value of the RFR model is 0.9826, and the actual measured curvature value is highly correlated with the model predicted curvature value. Compared to that exhibited by back‐propagation neural networks, the RFR model exhibits higher accuracy for curvature estimation, with average values of 0.1314 and 0.1173 for root mean square and mean absolute errors, respectively. This method can provide a more comprehensive theoretical basis for the application of robot learning in the curvature measurement of LPFG and has practical value.

At present, the trend of population aging in the world is becoming more and more serious. Many countries have implemented many measures to solve this problem. Various high-tech products to help the elderly and disabled have also entered the lives of the elderly. In order to ensure the safety of the elderly or disabled at home, this paper designs a room using SLAM algorithm of lidar sensor under ROS (Robot Operating System) environment. Wheeled home service robots with internal positioning, mapping and multi-point continuous navigation can collect and analyze environmental data such as temperature, humidity, smoke and rainwater outside the windows and broadcast them by voice while navigation. After many debugging, this design has realized the function of indoor autonomous navigation, and timely alarm and control of abnormal environmental data in the home.

Human action recognition is an important part of intelligent video processing, which has high research value and broad application prospects. This paper takes the video monitoring of the elderly living alone as an example to recognize human behavior, and chooses the video of indoor background taken by a single camera as the research object. The background subtraction method is used to extract moving objects from video, and the extracted moving objects are pretreated. Then, according to the overall outline of the object or the obvious features of a part of the object, the motion features are obtained to understand and identify the movement of the monitored object, such as walking and falling. Experiments show that the proposed algorithm has better processing effect for actual video.