Tunable bandwidth terahertz perfect absorption device based on vanadium dioxide phase transition control.

Utilizing the phase transition principle of VO2, this paper presents a tunable ultra-wideband terahertz perfect absorption device with simple structure and tunability. The proposed broadband terahertz perfect absorption device is a three-layer structure with a metal reflective layer, a silicon dioxide dielectric layer and a VO2 layer from bottom to top. It was found that the terahertz perfect absorption device's absorption could be dynamically adjusted from 1.2% to 99.9% when changing from an insulated to a metallic state. With the VO2 in the metallic state, the terahertz perfect absorption device has an absorption efficiency of more than 90% in 4.00 to 10.08 THz's ultra-broadband range and near-perfect absorption is achieved in the ranges of 4.71 THz to 5.16 THz and 7.74 THz to 8.06 THz. To explain the working principle of this terahertz perfect absorption device, this paper utilizes wave interference's principle, theory of impedance matching and electric field analysis. Compared to previously reported terahertz metamaterial devices, the vanadium dioxide device proposed in this paper is significantly optimized in terms of tunable range and absorption bandwidth. In addition, the terahertz perfect absorption device is polarization insensitive and maintains good absorptivity over a wide-angle incidence range. This tunable ultra-wideband terahertz perfect absorption device could have applications in the fields of modulation, stealth devices, and thermal emission devices.

The Effect of Height Error on Performance of Propagation Phase-Based Metalens.

Metalenses, as a new type of planar optical device with flexible design, play an important role in miniaturized and integrated optical devices. Propagation phase-based metalenses, known for their low loss and extensive design flexibility, are widely utilized in optical imaging and optical communication. However, fabrication errors introduced by thin-film deposition and etching processes inevitably result in variations in the height of the metalens structure, leading to the fabricated devices not performing as expected. Here, we introduce a reflective TiO2 metalens based on the propagation phase. Then, the relationship between the height variation and the performance of the metalens is explored by using the maximum phase error. Our results reveal that the height error of the unit structure affects the phase rather than the amplitude. The focusing efficiency of our metalens exhibits robustness to structural variations, with only a 5% decrease in focusing efficiency when the height varies within ±8% of the range. The contents discussed in this paper provide theoretical guidance for the unit design of the propagation phase-based metalens and the determination of its allowable fabrication error range, which is of great significance for low-cost and high-efficiency manufacturing.

Simultaneous Passivation of Perovskite Interfaces at Multiple Active Sites Improves Device Performance: Combining Theory and Experiment.

A multisite interface passivation material named 2-mercapto-4-methyl-5-thiazoleacetic acid (MMTA) is used to optimize the perovskite film top interface. DFT calculations and experiments show that MMTA can effectively passivate interface defects. Finally, the champion device's photoelectric conversion efficiency reached 23.44%, which possessed long-term stability.

Grasping detection of dual manipulators based on Markov decision process with neural network.

With the development of artificial intelligence, robots are widely used in various fields, grasping detection has been the focus of intelligent robot research. A dual manipulator grasping detection model based on Markov decision process is proposed to realize the stable grasping with complex multiple objects in this paper. Based on the principle of Markov decision process, the cross entropy convolutional neural network and full convolutional neural network are used to parameterize the grasping detection model of dual manipulators which are two-finger manipulator and vacuum sucker manipulator for multi-objective unknown objects. The data set generated in the simulated environment is used to train the two grasping detection networks. By comparing the grasping quality of the detection network output the best grasping by the two grasping methods, the network with better detection effect corresponding to the two grasping methods of two-finger and vacuum sucker is determined, and the dual manipulator grasping detection model is constructed in this paper. Robot grasping experiments are carried out, and the experimental results show that the proposed dual manipulator grasping detection method achieves 90.6% success rate, which is much higher than the other groups of experiments. The feasibility and superiority of the dual manipulator grasping detection method based on Markov decision process are verified.

Ultra-Wideband High-Efficiency Solar Absorber and Thermal Emitter Based on Semiconductor InAs Microstructures.

Since the use of chemical fuels is permanently damaging the environment, the need for new energy sources is urgent for mankind. Given that solar energy is a clean and sustainable energy source, this study investigates and proposes a six-layer composite ultra-wideband high-efficiency solar absorber with an annular microstructure. It achieves this by using a combination of the properties of metamaterials and the quantum confinement effects of semiconductor materials. The substrate is W-Ti-Al2O3, and the microstructure is an annular InAs-square InAs film-Ti film combination. We used Lumerical Solutions' FDTD solution program to simulate the absorber and calculate the model's absorption, field distribution, and thermal radiation efficiency (when it is used as a thermal emitter), and further explored the physical mechanism of the model's ultra-broadband absorption. Our model has an average absorption of 95.80% in the 283-3615 nm band, 95.66% in the 280-4000 nm band, and a weighted average absorption efficiency of 95.78% under AM1.5 illumination. Meanwhile, the reflectance of the model in the 5586-20,000 nm band is all higher than 80%, with an average reflectance of 94.52%, which has a good thermal infrared suppression performance. It is 95.42% under thermal radiation at 1000 K. It has outstanding performance when employed as a thermal emitter as well. Additionally, simulation results show that the absorber has good polarization and incidence angle insensitivity. The model may be applied to photodetection, thermophotovoltaics, bio-detection, imaging, thermal ion emission, and solar water evaporation for water purification.

Highly sensitive plasmonic sensor based on eccentric-core photonic crystal fibers.

To further reduce the fabrication difficulty of optical fiber sensors and improve the sensing performance, this study introduced the surface plasmon resonance (SPR) effect into optical fiber sensing technology and designed an eccentric-core photonic crystal fiber (EC-PCF). We investigated the characteristics of the two fundamental modes in the fiber core and the surface plasmon polariton (SPP) modes on the surface of the gold film. We also investigated the influence of the structural parameters, such as gold film coating area and thickness, air hole diameter, and eccentricity, on the confinement loss and achieved a refractive index (RI) sensitivity of 31.25 µm RIU-1 in the RI range of 1.29-1.43, corresponding to a figure of merit (FOM) of 521.6 per RIU. When the resolution of the optical spectrum analyzer was 0.1 nm, the EC-PCF could achieve a refractive index resolution of 3.2 × 10-6 RIU. Moreover, we performed tests with two typical sensing types, one in which the sensor was directly in contact with adulterated gasoline to achieve kerosene-concentration detection, and another in which the sensor was coated with a layer of polydimethylsiloxane (PDMS), whose RI is sensitive to the temperature field, to achieve temperature sensing. The EC-PCF demonstrated excellent sensing performance and offers obvious manufacturing advantages, providing a new and easily fabricated structural design idea for optical fiber sensing.

Dynamically changeable terahertz metamaterial absorbers with intelligent switch and high sensitivity and wide and narrow band perfect absorption.

In this work, we theoretically designed a dynamically changeable terahertz metamaterial absorber with intelligent switch and high sensitivity, wide band and narrow band perfect absorption based on the combination of Dirac semimetal (BDS) and vanadium dioxide (VO2). It features two methods for absorption adjustment: altering the Fermi energy level of BDS to modify the resonant frequency of the absorption peaks and utilizing the phase change of VO2 to regulate the absorption rate of the peaks. In addition, its rotational symmetric design ensures strong polarization-insensitivity. The simulation results demonstrate the presence of two narrowband absorption peaks and one mini-broadband absorption peak within the frequency range of 6.0-9.5 THz, all with absorption rates exceeding 90%. We provide an explanation of the absorption mechanism of the device, employing the relative impedance theory and localized surface plasmon resonance to analyze its electric field distribution. We also defined the refractive index sensitivity (S), which is SI = 378 GHz per RIU and SIII = 204 GHz per RIU. Our device possesses high sensitivity and two methods of adjusting absorption modes, which endow it with advantages in the fields of metamaterial absorbers, intelligent switch, and optical sensors.

Multi-view fusion network-based gesture recognition using sEMG data.

sEMG(surface electromyography) signals have been widely used in rehabilitation medicine in the past decades because of their non-invasive, convenient and informative features, especially in human action recognition, which has developed rapidly. However, the research on sparse EMG in multi-view fusion has made less progress compared to high-density EMG signals, and for the problem of how to enrich sparse EMG feature information, a method that can effectively reduce the information loss of feature signals in the channel dimension is needed. In this paper, a novel IMSE (Inception-MaxPooling-Squeeze- Excitation) network module is proposed to reduce the loss of feature information during deep learning. Then, multiple feature encoders are constructed to enrich the information of sparse sEMG feature maps based on the multi-core parallel processing method in multi-view fusion networks, while SwT (Swin Transformer) is used as the classification backbone network. By comparing the feature fusion effects of different decision layers of the multi-view fusion network, it is experimentally obtained that the fusion of decision layers can better improve the classification performance of the network. In NinaPro DB1, the proposed network achieves 93.96% average accuracy in gesture action classification with the feature maps obtained in 300ms time window, and the maximum variation range of action recognition rate of individuals is less than 11.2%. The results show that the proposed framework of multi-view learning plays a good role in reducing individuality differences and augmenting channel feature information, which provides a certain reference for non-dense biosignal pattern recognition.

Active thermally tunable and highly sensitive terahertz smart windows based on the combination of a metamaterial and phase change material.

A thermally tunable terahertz window based on the combination of a metamaterial and the phase change material VO2 is proposed. The window is composed of two vanadium oxide films with a SiO2 layer sandwiched between them. The thermochromic phase change properties of VO2 are the key to the functionality of the window. By controlling the temperature around the room temperature of 300 K, our material can be used as a smart window and it is able to regulate both the absorption and transmission of external terahertz waves in response to changes in temperature. The absorbance can be regulated by more than 90% and the transmittance by more than 80%. The switching characteristics of the window are explained by the insulator-metal transition that vanadium oxide undergoes during the heating process, while localized surface plasmon resonance explains the perfect absorption. In addition, the designed window is not only insensitive to polarised waves, but is thermally flexible and maintains excellent performance over a wide angular range of 0° to 40°. This design will have significant potential for applications in stealth technologies, thermal sensing and switching, and terahertz energy harvesting.

High confidence plasmonic sensor based on photonic crystal fibers with a U-shaped detection channel.

In order to improve the performance of optical fiber sensing and expand its application, a photonic crystal fiber (PCF) plasmonic sensor with a U-shaped channel based on surface plasmon resonance (SPR) is proposed. We have studied the general influence rules of structural parameters such as the radius of the air hole, the thickness of the gold film and the number of U-shaped channels using COMSOL based on the finite element method. The dispersion curves and loss spectrum of the surface plasmon polariton (SPP) mode and the Y-polarization (Y-pol) mode as well as the distribution of the electric field intensity (normE) under various conditions are studied using the coupled mode theory. The maximum refractive index (RI) sensitivity achieved in the RI range of 1.38-1.43 is 24.1 µm RIU-1, which corresponds to a full width at half maximum (FWHM) of 10.0 nm, a figure of merit (FOM) of 2410 RIU-1 and a resolution of 4.15 × 10-6 RIU. The results show that the proposed sensor combines the SPR effect, which is extremely sensitive to changes in the RI of the surrounding medium and realizes real-time detection of the external environment by analyzing the light signal modulated by the sensor. In addition, the detection range and sensitivity can be extended by adjusting the structural parameters. The proposed sensor has a simple structure with excellent sensing performance, which provides a new idea and implementation method for real-time detection, long-range measurement, complex environment monitoring and highly integrated sensing, and has a strong potential practical value.

Behavior Strategy Analysis Based on the Multi-Stakeholder Game under the Plastic Straw Ban in China.

Since 1 January 2021, China has banned nondegradable disposable straws in the catering industry. To promote the enforcement of the ban of plastic straws and improve the relationship between economic development and environmental protection, based on the evolutionary game method, this paper constructs the game model from the supply side and the demand side, respectively. Subsequently, through the dynamic equation, stable system evolution strategy is obtained. Furthermore, simulation is conducted to test the influence of the main parameters in the model on the evolution of system strategy. The results show that (1) the change of the government strategy mainly depends on its regulation costs and revenue, while the production strategy of a company is affected by the government and consumer strategies. (2) From the perspective of enterprise supply, government subsidies can promote technological innovation and develop new plastic straw substitutes. However, government penalties have little effect on violating enterprises. In addition, from the perspective of enterprise demand, with the collaboration of enterprises and consumers, it is easier for enterprises to carry out technological innovation. (3) Consumer acceptance of the substitutes for disposable plastic straws as well as online comments have a decisive influence on the enterprises' selections for research and development (R&D) strategies.

A Time Sequence Images Matching Method Based on the Siamese Network.

The similar analysis of time sequence images to achieve image matching is a foundation of tasks in dynamic environments, such as multi-object tracking and dynamic gesture recognition. Therefore, we propose a matching method of time sequence images based on the Siamese network. Inspired by comparative learning, two different comparative parts are designed and embedded in the network. The first part makes a comparison between the input image pairs to generate the correlation matrix. The second part compares the correlation matrix, which is the output of the first comparison part, with a template, in order to calculate the similarity. The improved loss function is used to constrain the image matching and similarity calculation. After experimental verification, we found that it not only performs better, but also has some ability to estimate the camera pose.

Dynamic Gesture Recognition Algorithm Based on 3D Convolutional Neural Network.

Gesture recognition is one of the important ways of human-computer interaction, which is mainly detected by visual technology. The temporal and spatial features are extracted by convolution of the video containing gesture. However, compared with the convolution calculation of a single image, multiframe image of dynamic gestures has more computation, more complex feature extraction, and more network parameters, which affects the recognition efficiency and real-time performance of the model. To solve above problems, a dynamic gesture recognition model based on CBAM-C3D is proposed. Key frame extraction technology, multimodal joint training, and network optimization with BN layer are used for making the network performance better. The experiments show that the recognition accuracy of the proposed 3D convolutional neural network combined with attention mechanism reaches 72.4% on EgoGesture dataset, which is improved greatly compared with the current main dynamic gesture recognition methods, and the effectiveness of the proposed algorithm is verified.

Single-celled multifunctional metasurfaces merging structural-color nanoprinting and holography.

Nanostructured metasurfaces applied in structural-color nanoprinting and holography have been extensively investigated in the past several years. Recently, merging them together is becoming an emerging approach to improve the information capacity and functionality of metasurfaces. However, current approaches, e.g., segmenting, interleaving and stacking schemes for function merging, suffer from crosstalk, low information density, design and fabrication difficulties. Herein, we employ a single-celled approach to design and experimentally demonstrate a high-density multifunctional metasurface merging nanoprinting and holography, i.e., each nanostructure in the metasurface can simultaneously manipulate the spectra (enabled with varied dimensions of nanostructures) and geometric phase (enabled with varied orientation angles of nanostructures) of incident light. Hence, with different decoding strategies, a structural-color nanoprinting image emerges right at the metasurface plane under white light illumination, while a holographic image is reconstructed in the Fraunhofer diffraction zone under circularly polarized laser light incidence. And the two images have no crosstalk since they are independently designed and presented at different distances. Our proposal suggests a space-multiplexing scheme to develop advanced metasurfaces and one can find their markets in high-density information storage, optical information encryption, multi-channel image display, etc.

Full-space metasurface holograms in the visible range.

Conventional metasurface holography is usually implemented in either transmission space or reflection space. Herein, we show a dielectric metasurface that can simultaneously project two independent holographic images in the transmission and reflection spaces, respectively, merely with a single-layer design approach. Specifically, two types of dielectric nanobricks in a nanostructured metasurface are employed to act as half-wave plates for geometric phase modulation. One type of nanobrick is designed to reflect most of incident circularly-polarized light into reflection space, enabled with magnetic resonance, while another type of nanobrick transmits it into transmission space, without resonance involved. By controlling the orientation angles and randomly interleaving the two types of nanobricks to form a metasurface, a full-space metasurface hologram can be established. We experimentally demonstrate this trans-reflective meta-holography by encoding the geometric phase information of two independent images into a single metasurface, and all observed holographic images agree well with our predictions. Our research expands the field-of-view of metasurface holography from half- to full-space, which can find its markets in optical sensing, image displays, optical storages and many other potential applications.

On-axis three-dimensional meta-holography enabled with continuous-amplitude modulation of light.

Conventional three-dimensional (3D) holography based on recording interference fringes on a photosensitive material usually has unavoidable zero-order light, which merges with the holographic image and blurs it. Off-axis design is an effective approach to avoid this problem; however, it in turn leads to the waste of at least half of the imaging space for holographic reconstruction. Herein, we propose an on-axis 3D holography based on Malus-assisted metasurfaces, which can eliminate the zero-order light and project the holographic image in the full transmission space. Specifically, each nanostructure in the metasurface acts as a nano-polarizer, which can modulate the polarization-assisted amplitude of incident light continuously, governed by Malus law. By carefully choosing the orientation angles of nano-polarizers, the amplitude can be both positive and negative, which can be employed to extinct zero-order light without affecting the intensity modulation for holographic recording. We experimentally demonstrate this concept by projecting an on-axis 3-layer holographic images in the imaging space and all experimental results agree well with our prediction. Our proposed metasurface carries unique characteristics such as ultracompactness, on-axis reconstruction, extinction of zero-order light and broadband response, which can find its market in ultracompact and high-density holographic recording for 3D objects.

Learning for a Robot: Deep Reinforcement Learning, Imitation Learning, Transfer Learning.

Dexterous manipulation of the robot is an important part of realizing intelligence, but manipulators can only perform simple tasks such as sorting and packing in a structured environment. In view of the existing problem, this paper presents a state-of-the-art survey on an intelligent robot with the capability of autonomous deciding and learning. The paper first reviews the main achievements and research of the robot, which were mainly based on the breakthrough of automatic control and hardware in mechanics. With the evolution of artificial intelligence, many pieces of research have made further progresses in adaptive and robust control. The survey reveals that the latest research in deep learning and reinforcement learning has paved the way for highly complex tasks to be performed by robots. Furthermore, deep reinforcement learning, imitation learning, and transfer learning in robot control are discussed in detail. Finally, major achievements based on these methods are summarized and analyzed thoroughly, and future research challenges are proposed.

Structural-color nanoprinting with hidden watermarks.

Nanostructured metasurfaces can manipulate the spectrum and polarization of incident light at the nanoscale, which suggests a new integration of color nanoprints and polarizing-related components. Herein, we design and experimentally demonstrate a structural-color nanoprint carrying hidden watermarks, enabled with the polarization-assisted spectrum manipulation of light. Specifically, under unpolarized white light, the watermarks are concealed and a structural-color nanoprinting-image occupies the metasurface plane. Meanwhile, once linearly polarized white light is incident on the same metasurface, the hidden information can be decoded, and the same nanoprinting-image covered with watermarks appears. The proposed metasurface represents a paradigm for displaying color nanoprinting-images with or without watermarks, showing a flexible switch between the two operating modes and providing an easily camouflaged scheme for anticounterfeiting, encryption, information multiplexing, high-density optical storage, etc.

Combining Public Opinion Dissemination with Polarization Process Considering Individual Heterogeneity.

The wide dissemination of false information and the frequent occurrence of extreme speeches on online social platforms have become increasingly prominent, which impact on the harmony and stability of society. In order to solve the problems in the dissemination and polarization of public opinion over online social platforms, it is necessary to conduct in-depth research on the formation mechanism of the dissemination and polarization of public opinion. This article appends individual communicating willingness and forgetting effects to the Susceptible-Exposed-Infected-Recovered (SEIR) model to describe individual state transitions; secondly, it introduces three heterogeneous factors describing the characteristics of individual differences in the Jager-Amblard (J-A) model, namely: Individual conformity, individual conservative degree, and inter-individual relationship strength in order to reflect the different roles of individual heterogeneity in the opinions interaction; thirdly, it integrates the improved SEIR model and J-A model to construct the SEIR-JA model to study the formation mechanism of public opinion dissemination and polarization. Transmission parameters and polarization parameters are simulated and analyzed. Finally, a public opinion event from the pricing of China's self-developed COVID-19 vaccine are used, and related Weibo comment data about this event are also collected so as to verify the rationality and effectiveness of the proposed model.

Non-orthogonal-polarization multiplexed metasurfaces for tri-channel gray-imaging.

Metasurface based polarization multiplexing is usually conducted in two orthogonal-polarization states, e.g., linearly polarized along x/y axes, left/right-handed circularly polarized states, etc. Herein, we show metasurfaces can be employed to implement tri-channel polarization multiplexing in three non-orthogonal-polarization states, merely with a single-size nanostructure design approach. Specifically, nanostructured metasurfaces acting as nano-polarizer arrays can modulate the incident light intensity pixel-by-pixel by controlling the orientation angles of nanostructures, governed by Malus's law. Hence, by inserting a metasurface between a bulk-optic polarizer and an analyzer, and elaborately controlling their polarization combinations, we show that the Malus-assisted metasurface can simultaneously record a continuous gray-image and two independent binary-patterns in three different information channels. We experimentally demonstrate this concept by recording three independent gray-images right at the metasurface surface. With the advantages of high information density, high security, high compatibility and ultracompactness, the proposed gray-imaging meta-device can play a significant role in the field of optical storage, anti-counterfeiting, and information multiplexing, etc.