Search

The COVID-19 outbreak has caused the mortality worldwide and the use of swab sampling is a common way of screening and diagnosis. To combat respiratory infectious diseases and assist sampling, robots have been utilized and shown promising potentials. Nonetheless, a safe, patient-friendly, and low-cost swabbing system would be crucial for the practical implementation of robots in hospitals or inspection stations. In this study, we proposed two recyclable and cost-efficient end-effector designs that can be equipped at the distal end of a robot to passively regulate or actively sense the force exerted onto patients. One way is to introduce passive compliant mechanisms with soft material to increase the flexibility of the swabbing system, while the other way is utilizing a force-sensing gripper with embedded optoelectronic sensors to actively sense the force or torque. The proposed designs were modelled computationally and tested experimentally. It is identified that the passive compliant mechanisms can increase the flexibility of the swabbing system when subjected to the lateral force and mitigate the vertical force resulted from buckling. The lateral force range that the force-sensing gripper can detect is 0-0.35 N and the vertical force range causing buckling effect that can be sensed by gripper is 1.5-2.5 N.

Research on emotion recognition that relies purely on eye-tracking data is very limited although the usability of eye-tracking technology has great potential for emotional recognition. This paper proposes a novel approach for 4-class emotion classification using eye-tracking data solely in virtual reality (VR) with machine learning algorithms. We classify emotions into four specific classes using VR stimulus. Eye fixation data was used as the emotional-relevant feature in this investigation. A presentation of 3600 videos, which contains four different sessions, was played in VR to evoke the user’s emotions. The eye-tracking data was collected and recorded using an add-on eye-tracker in the VR headset. Three classifiers were used in the experiment, which are k-nearest neighbor (KNN), random forest (RF), and support vector machine (SVM). The findings showed that RF has the best performance among the classifiers, and achieved the highest accuracy of 80.55%.

For a long time, the research on the structure design and gait control of the curved leg hexapod robot has been more in-depth, but there is less research on the energy consumption during the movement. At the same distance, legged robots’ energy consumption is significantly higher than that of wheeled robots of the same size, so reducing energy consumption can increase the robot’s endurance. In this paper, for flat terrain, based on dynamic analysis, polynomial, cosine, and trapezoidal velocity motion curves are planned, separately. At the same time, we obtain the motion curves based on the minimum energy consumption optimization. In the experiment and simulation analysis, the influence of gait parameters such as different curve trajectories, support phases, and periods on energy consumption are discussed separately. The simulation and experimental results show that under the same motion period, the larger the support phase angle, the more energy consumption; under the same support phase, the energy consumption first decreases and then increases as the period increases; the energy consumption of the optimized curve is reduced by 22.3%, 29.5%, and 15.3% respectively compared with the energy consumption of the trapezoidal speed movement curve. The results of this paper provide a useful reference for the curved leg hexapod robot in energy-saving motion.

A wheel-legged mobile robot platform, namely Land Devil Ray-2 (LDR-2), is proposed with adjustable body length. To obtain the influence of the parameters of the mechanism on the motion, the relationship between the length of the robot and the stability of obstacle crossing is analyzed. The robot platform builds a robot perception system based on cameras and other sensing devices, performing tasks such as exploration or search and rescue in complex terrain. The robot control system can control the robot to switch motion modes and extend and retract the tail according to obstacle parameters. After completing the analysis and design, a robot prototype is built, and its obstacle-crossing performance is tested. The experiment results show that the proposed robot significantly improves the obstacle crossing performance than similar wheel-legged robots. This method can perform tasks better in different working environments and has reference value for the design of search and rescue mobile robots.

Amphibious robots can be used in both land and underwater environments, which has a wider range of applications than traditional mobile robots. However, the development of propulsion mechanisms for use in amphibious environments is very challenging. In this study, a new type of composite propulsion mechanism is presented, hinged by a plurality of deformable joints, and is suitable for amphibious environments. By changing the relative position of the rotation center of the slider chain and the rotation center of the outer track, the mechanism can be transformed between the land curved-legged state and the underwater webbed state. The mechanism has the advantages of high terrain adaptability and fast speed of curved legs when moving on land, and has good webbed flexibility when moving underwater. In this paper, aiming at the typical terrain of the amphibious environment, a new amphibious robot equipped with the propulsion mechanism is developed, and the kinematic performance of the amphibious robot is tested. The results show that the design method of composite propulsion mechanism suitable for the amphibious environment can improve amphibious robots’ kinematic performance and provide a valuable reference for the design and control of other amphibious robots.

Motivated by the correlation between the depth and the geometric structure of a 360 indoor image, we propose a novel learning-based depth estimation framework that leverages the geometric structure of a scene to conduct depth estimation. Specifically, we represent the geometric structure of an indoor scene as a collection of corners, boundaries and planes. On the one hand, once a depth map is estimated, this geometric structure can be inferred from the estimated depth map; thus, the geometric structure functions as a regularizer for depth estimation. On the other hand, this estimation also benefits from the geometric structure of a scene estimated from an image where the structure functions as a prior. However, furniture in indoor scenes makes it challenging to infer geometric structure from depth or image data. An attention map is inferred to facilitate both depth estimation from features of the geometric structure and also geometric inferences from the estimated depth map. To validate the effectiveness of each component in our framework under controlled conditions, we render a synthetic dataset, Shanghaitech-Kujiale Indoor 360 dataset with 3550 360 indoor images. Extensive experiments on popular datasets validate the effectiveness of our solution. We also demonstrate that our method can also be applied to counterfactual depth.

Emotional analysis in literary and artistic art is a current research hotspot, while emotional-based fiction classification is more complicated than other literary forms. Emotional detection is usually done from the perspective of voice or facial images. Most of the existing text-based research uses typical binary or ternary classification methods. Therefore, how to extract information of emotion from the novel text and classify it is still a problem to be solved. In this paper, a Bi-directional Long Short-Term Memory Language Model (BiLSTM-LM) is proposed. The text sequence is divided into six different emotional categories. Word Embedding method is used to encode vocabulary, and the text is represented as word feature representation and character feature representation respectively. The final emotional output is estimated by combining content words and emotional function words. Experimental results show that the training of the model in this paper can quickly converge, and the detection accuracy is up to 64.09%, which is 4 percentage points higher than the current best model, and basically can effectively classify novels.

To simultaneously estimate head counts and localize heads with bounding boxes, a regression guided detection network (RDNet) is proposed for RGB-D crowd counting. Specifically, to improve the robustness of detection-based approaches for small/tiny heads, we leverage density map to improve the head/non-head classification in detection network where density map serves as the probability of a pixel being a head. A depth-adaptive kernel that considers the variances in head sizes is also introduced to generate high-fidelity density map for more robust density map regression. Further, a depth-aware anchor is designed for better initialization of anchor sizes in detection framework. Then we use the bounding boxes whose sizes are estimated with depth to train our RDNet. The existing RGB-D datasets are too small and not suitable for performance evaluation on data-driven based approaches, we collect a large-scale RGB-D crowd counting dataset. Experiments on both our RGB-D dataset and the MICC RGB-D counting dataset show that our method achieves the best performance for RGB-D crowd counting and localization. Further, our method can be readily extended to RGB image based crowd counting and achieves comparable performance on the ShanghaiTech Part\_B dataset for both counting and localization.

Aiming at the problem that uncertain power sources bring fluctuation to power system and cause lots of wind and photovoltaic curtailment, a short-term optimized operation method of multi-energy power system based on complementary characteristics of power sources is presented. The complementary mechanism of multi-energy power system is established. Based on the complementary mechanism, mathematical models of complementary objectives and complementary demands are established. Define renewable energy complementary power source. To minimize complementary demands, the capacity matching ratio of hydropower in renewable energy complementary power source is determined. According to the complementary objectives, a hierarchical optimal operation strategy for multi-energy power system is proposed. Particle swarm optimization is used to solve the optimization objective function. Taking a provincial power grid in northwest China as an example, the results show that the strategy makes full use of the complementarity of power sources and has the benefit of energy saving.

Segmentation accuracy is critical in CBCT (cone-beam computed tomography) nondestructive detection. And it is influenced by the segmentation accuracy of CBCT serial slice images. However, the noise and artifacts in CBCT images make it hard to segment CBCT images precisely. To increase CBCT image segmentation accuracy, the 3D information in CBCT images should be fully used. We proposed and compared four connection models for CBCT images pretreatment. They can decrease the noise in CBCT images. Moreover, we propose a 3D CBCT image segmentation method based on the accumulated FCM_S. In the experiment, CBCT slice images of a workpiece are segmented by our proposed method and comparing methods. The segmentation results certified the effectiveness of our method.

A new Gaussian-fast median filtering algorithm is proposed to filter the image of tunnel cracks in view of the shortcomings of the traditional method of filtering the cracks in the subway tunnel. The algorithm calculates the weighted average value for the image filter template window, and compares the pixel gray value of each point in the image to the weighted mean value. If the pixel gray value of the point is greater than its weighted average value, the fast algorithm of median filter algorithm is used to process it, otherwise the gauss filter algorithm is used to process it. By comparing with median and gauss filtering results, the new algorithm proposed effectively reduces the image noise and improves the image quality, effectively protects the edge of image cracks, shortens the processing time and improves the efficiency greatly, and provides the foundation for subsequent crack recognition and segmentation. The simulation results show that the proposed algorithm has a good balance between the processing speed and the effect. The effectiveness of the proposed algorithm is verified by the same segmentation method.

In this paper, a rule-based method is proposed to segment optic disc. Firstly, the results of color-boosted Harris corner detection are applied to calculate the convex hull which represents an approximate region of the optic disc. Secondly, two super-pixel methods are used to over segment the retinal image separately, based on which the prior probability and the observation likelihood probability of the constructed Bayesian model are calculated to estimate the posterior probability of the optic disc. Thirdly, the posterior probability using Otsu threshold segmentation method is binarized, and finally the accurate segmentation result of optic disc using Hough transform is obtained. A large number of experimental results verify the effectiveness of the proposed algorithm.

T139 DDR4 SDRAM is the latest Double-Data-Rate Fourth Generation Synchronous Dynamic Random Access Memory with the advantages of large capacity, high speed, low power consumption and good stability. In order to solve the data cache problem of the ultra-high-speed sampling and processing system, this paper designs DDR4 SDRAM read-write controller module based on UltraScale DDR4_v2.2 IP core. This paper mainly introduces the characteristics and working principle of DDR4, introduces the DDR4 read-write controller module in detail, and packages it as the FIFO interface and finally tests and gives the test results. System tests results verify that the module meets the requirements of large-capacity data cache, and has the advantages of good stability and high portability.

In order to monitor series arc fault in real-time for electrical connectors and improve the reliability of power supply systems, series arc fault experiments were carried out using an arc fault generator. A three-phase asynchronous motor and a three-phase frequency conversion motor were used as experimental loads. The variance, covariance and number of zero-crossing points of five adjacent periods of current signals were extracted and normalized. The feature vector was constructed by using the above variables such as number of zero-crossing points, variance and covariance. The k-nearest neighbor method was used for pattern recognition of the feature vector. The results showed that this method was effective for the diagnosis of series arc fault in electrical connectors.

In this paper, we investigate on constructing cryptographically significant Boolean functions with n variables based on decompositions of the multiplicative group of the finite field F 2n of the form F 2n * = U × V, where U and V are cyclic subgroups of F 2n * satisfying (|U|, |V|) = 1. For positive integers s, m and n = 2sm, we obtain classes of unbalanced functions with optimal algebraic immunity in the cases |U| = 2m + 1, |V| = (2n−1)/(2m+1) and |U| = 2m−1, |V| = (2n−1)/(2m−1), respectively, where in the latter case the optimal algebraic immunity is based on correctness of the Tu-Deng conjecture. Functions belonging to both classes can be modified to be balanced ones with (potentially) optimal algebraic immunity and optimal algebraic degree, and computer experiments show that they also have high nonlinearity and good immunity against fast algebraic attacks. As by-products, variants of the Tu-Deng conjecture and combinatorial results on binary strings in analogy to it are also obtained.

Power system operation is confronted with the problem of how to reduce the damage that caused by the ice on the transmission lines as much as possible. Due to the indeterminacy of icing thickness on the transmission lines, accurate prediction of the icing thickness is essential in order to effectively guide the transmission line route planning and ice-resistant design of electric grid. Currently, there are relatively few research results on the medium and long term prediction of the icing thickness on the transmission lines. This paper presents a transmission line icing thickness prediction model that based on fuzzy Markov chain prediction, combining with the practical measured data from the Czech Studnice Observatory between 1940 and 1999. The model was verified through forecasting the range of the annual maximum of icing thickness during 1994-1999. The result shows that the accuracy of this method reached 80%.

This paper proposes an improved structural similarity method for audio quality assessment, which is Structural Similarity Analysis of Modulation (SSAM). Different from original structural similarity index measure, we introduce the analysis of structural similarity of modulation together with Computational Auditory Signal-processing and Perception (CASP) model. Audio features from CASP are extended to three dimensions: time, frequency and modulation spectrum. The combined architecture of ITU-R BS.1387-1 and proposed SSAM is given in this paper. Our proposed estimation system not only shows highly correlated with subjective results but also overcomes the shortage of ITU-R BS.1387-1 that only suitable for small impaired audio.

Measurement of audio quality is of fundamental importance for numerous speech and audio processing applications, where the goal of quality assessment algorithms is to evaluate the quality in agreement with human's subjective feelings of audio quality. This paper includes a brief summary of the development of intrusive and non-intrusive audio quality assessment algorithms and outlines recent advancements, especially some new topics to be investigated presently. Moreover, according to the existing problems in this field, discussion of the future trend and challenges is also given.

Currently, many countries are seeking to develop new energy vehicles. AC charge spots are the major energy supply facilities for family electric vehicle, which use the on-board charger. However, due to the non-linear character of on-board charger, the harmonic current will be injected into the grid, and will affect the grid quality within the large scale charge. This paper explores that the active power filter is applied to the AC charge spots, in order to suppress the harmonics of on-board charger which directly interferes with power grids. Applying to two different non-linear loads, the experimental prototype of the single-phase parallel active filter and the composite control method can effectively restrain the harmonics of the on-board charger in order to improve the charging efficiency.

This paper proposed an improved audio quality metric via circular and flexible overlap. Based on the Power Spectrum Estimation via circular overlap, we use a novel circular overlap sub-frame to assess highly impaired audio. The relationship between the fraction of overlap and the metric of audio quality is also examined, and it has been proven that the accuracy of audio quality assessment increased with the fraction of overlap. By integrating circular and flexible overlap into ITU-R BS.1387, which is also known as Perceptual Evaluation of Audio Quality, our method can be applied to the quality assessment of highly impaired audio.

Based on the evolution of the brand community, this paper constructed a theoretical model of industrial clusters' development and evolution in order to explore the pattern of fostering regional clusters brands and famous corporate brand, turn the traditional industrial cluster into a brand incubator and achieve positive self circulation and outer circulation.

In recent years, the Structure Design Competition in many colleges has made significant achievements. With experience from participants in the grade of 2008 and 2009 of Zhejing University City College, this paper analyse the current situation of competition management. Concerning the limitations in organization and mechanism of Structure Association, it put forward practical organization, normalization of member system and leaders election, and implement of conference system, which promote the competition extraordinary.

Witkey pattern can not only create a new employment model, but also affect the education model. Witkey websites have great influence on education and students because they can help more and more students with studying and finding jobs as new network platforms.

Based on blend-learning and constructivism educational theory, comprehensive curriculum, assessment, and training resources can be designed to be perfect fit for college students who learn art and design so that students can be trained to use their brains more effectively.

With the development of credit market and changes in credit risk, many new quantitative assessment models and management strategies in the area of measurement and management. In this study, we discuss AHP for designing a credit assessment system to discriminate good creditors from bad ones.

This paper presents an opto-physiological model (OPM) to quantify the absorbance of multi-layered tissue in imaging photoplethysmography (IPPG). The approach used to generate such a model is to revise the path length of the Beer Lambert law through the Monte Carlo (MC) simulation of multi-layered tissue. The OPM can mathematically quantify the effect of optical properties on the absorbance of multilayered tissue. Subsequently, the absorbance measured from homogeneous, multi-layered tissue phantoms compares with model predictions. To this end, the model is validated to predict the widest range of experimental outcomes while maintaining the highest possible level of accuracy. This study brings a new approach to understand the principle of IPPG.

Earth science refers to a method and theory of using physical and chemical principles, etc, to explore underground minerals and study geological structures as well as one of important means by which people know and understand the earth by gravity, magnetic force, conductivity, earthquake, radioactivity and logging, etc. The scientific data produced in earth science researches is called earth data. Earth data is an important source of earth science innovation as well as one of information data necessary for modern human economy, society, military affairs and daily activities. When exploring and solving practical problems by earth science method, we often need to combine it with various factors, such as geographical environment and geological condition, etc, to analyze and solve problems from many aspects and angles instead of using a certain method and means in isolation to solve problems one-sidedly. And therefore, establishing an earth science data sharing platform plays a very important role in the task of national space information infrastructure construction.