This PDF file contains the front matter associated with SPIE Proceedings Volume 12612, including the Title Page, Copyright information, Table of Contents, and Conference Committee listings.

In this paper, the finite element method is used to analyze the stress of the guide arm support under three different working conditions, and the equivalent stress nephogram and deformation nephogram of the guide arm support are obtained. The response surface optimization design module in Ansys Workbench is used to optimize the topology of the guide arm bracket. According to the optimization results, new materials are used to design and manufacture the guide arm bracket. The weight of the optimized guide arm bracket is reduced by 4.6943Kg, and the weight loss rate is 23.1%.

With the development of industrial design, traditional requirements of reasonable function and aesthetic appearance are insufficient for nowadays customers' needs, which leads to the proposition of Kansei Engineering (KE). This paper introduces Kansei and six types of KE methods, and provides a design application based on Hybrid KE system and computer data analysis to evaluate and improve an intelligent home nursing bed by principal component analysis and cluster analysis, of which accurately and efficiently reflected the performance of the design scheme in six pairs of key indicators. The open challenges and future directions of KE are discussed as well, suggestion of its future study is offered for better knowledge and quick understanding of KE by relevant scholars.

Aiming at the emerging issues of Chinese aging population in the context of social digitization, the elderly behaviours and habits when using applications on smart phone were studied in order to understand the inconvenience and obstacle they often encounter. Through mining and analyzing scatter data obtained by capturing the gaze areas and touch locations on the screen, the method of nested square data statistics was used in different modules of user interface. Data fit and variance analysis were conducted by utilizing MatLab and probabilistic models of Mathematical Statistics, which could set up the behaviours mathematical model of elderly user when using mobile applications. Finally, a principle of "senior mode" user interface design for mobile applications was proposes.

The objective of this paper is to provide a systematic view of how Participatory Design(PD) has been spread and developed over the last two decades(1998-2021) and to identify prospective challenges and opportunities. Both quantitative scientometrics analysis and qualitative analysis were adopted to identify focuses and trends in PD. With Web of Science (WoS) core collection as the data source, we used CiteSpace and VOS viewer software to visualize scholarly contribution, impact, and connections by analyzing the definition, methodology, and practice of PD. A theme-based qualitative analysis was used to reveal the evolution of participatory design research development and explores future trends.

China's industrial design teaching about ergonomics course is generally unable to let the students feel various products and human body size data relations intuitively. The users in this study are mainly teachers and students majoring in industrial design, which can be used as auxiliary teaching courseware for teachers. Meanwhile, this study hopes to improve the understanding and application ability of students majoring in industrial design on ergonomics through the design of innovative ergonomics teaching methods. Based on analysis of the ergonomics textbooks of Chinese college students, this study finally selected furniture design as the target content of the design and focused the design and interactive content of the virtual scene on the design of the home scene. First, the virtual reality technology application required the computer software Unity3D to build and render the home scene model in the scene. Then, representative furniture was selected through the analysis of the size and design points, and the knowledge point was added to the scene through an illustrative interface. Finally, a desktop application that could enable users to explore and learn from the virtual scene by wearing VR glass was generated, to enhance the interactive teaching and improve students' learning initiative.

In the post-epidemic era, more and more people are participating in physical exercise and increasing the frequency of outdoor activities. Outdoor folding chairs are favored by young consumers in outdoor activities because of their easy portability, space saving and flexible use scenarios. The purpose of this study is to dig deeper into the needs of target users for outdoor folding chairs, combine the product characteristics of outdoor folding chairs, and refine the design positioning to improve the user experience. First, the needs of the target users were obtained through desktop research, user journey maps, and user interviews, and the KJ method was used for hierarchical analysis. Then, the user needs were clustered according to the KANO model to discover the type and ranking of user needs. Finally, the KANO model and the QFD model are integrated for constructing the requirements and characteristics model, and the key design elements are derived from it for in-depth design. The proposed KANO-QFD feature combination design method enables designers to effectively grasp the real user requirements and grasp the key design elements to simplify the design process.

Sensory interaction in virtual technology can provide better immersive experience for users. In the field of design, the current virtual products are summarized from the perspectives of vision, hearing, touch, smell and multi-senses, and the hot topics and trends of the research are revealed by bibliometrics. The sensory interaction of the current virtual products is mainly from the interaction of vision with auditory, while the interaction of touch with olfactory is less. Literature research reveals that the research and application of VR is far more than that of other virtual technologies. Among the various senses, vision has been studied the most, followed by touch, hearing and multi-senses, while olfaction has been studied less. Main conclusion: More attention has been paid to the application of visual interaction in virtual products and literature research. Multi-sensory interaction is increasingly valued. The hot topics are virtual reality, human-computer interaction, environment and augmented reality, and keyword analysis indicate virtual environment, human-computer interaction, haptic interface, education are the future research trends.

The causes and treatments of autism are still a mystery to current research, but existing treatments have been tested and proven to be more effective. This article is a study of children with autism. By investigating the symptoms of autism and combining the current mainstream international treatment method ABA therapy, the toy design is innovated by combining the toys that are essential to every child's childhood with the treatment method, hope that the process of playing with toys can improve the core symptoms of autism. The article uses questionnaires, field research, face-to-face interviews and model testing to collate and summarise the data to demonstrate the feasibility of the design. After analysis, the author has designed a toy to help treat autistic children, while the toy design helps to alleviate the symptoms of autistic children to a certain extent, and provides some ideas for future therapeutic equipment design for autistic children.

In the past few years, the development of autopilot has made great progress. Autonomous vehicle is composed of many parts, including environmental perception and decision control. Steering angle prediction is particularly important in the control of autonomous vehicle and has attracted extensive attention. However, most of the existing models study the behavior methods during the day and have poor compatibility to night traffic environment. This paper is conducted to overcome this shortcoming through deep learning, which is supposed to automatically study relevant features from driving data without manual intervention. Specifically, the convolutional neural network (CNN) with residual structure has been used to build a model for learning and training, and the training data comes from the unity simulator. Based on this model, the steering wheel angle is predicted. The prediction results show that compared with other learning methods based on CNN, this model has higher accuracy in the case of poor lighting at night, and could be able to conform to several driving scenes as well as less revisions.

From the perspective of design development in smart textile installation, a focus on the performance indicators in experimenting wool felt in combination with laser cutting, fiber optics, and the performance indicators in experimenting woven mesh application and processes. To analytically theorise the combinations of unconventional fiber materials with automation technology and mechanical components in various forms of experience innovation.

Traditional rehabilitation therapy for children with autism (ASD) relies on manpower, is expensive, and lacks teaching aids. Based on augmented reality technology (AR), this research designs a game training software for children with ASD in terms of social cognition and motor perception, and combines the data monitoring module to record the children's usage in real time. The AHP-fuzzy comprehensive evaluation method is used to evaluate the software to verify the effectiveness of the software.

This paper aims to explore the application of scene theory in interaction design process of permission application platform, and establish a permission application platform design method model based on scene theory, to provide design and research ideas for designers and other researchers. Firstly, the problems existing in the game data permission application process of W company are analyzed and discovered. Then the application characteristics of scene theory are summarized in the design of the permission application platform, and the interaction design process model of the permission application platform is constructed under the guidance of scene theory. In the design practice of this process model, the business scenes and user scenes are sorted out and analyzed by using flowcharts. Finally, the validity and feasibility of the theoretical method are proved by designing the game data permission application platform of W company.

Objectve: The author intends to find out the best design scheme for a multi-function inflatable blanket for emergency rescue on the basis of the AHP/QFD analytical method. Method: The purpose of this study is to solve problems such as the singularity of designed functions of emergency rescue outfits currently available, as well as the lack of consideration for the rescuers from the aspect of the designer, etc. The author collects data through a questionnaire survey of actual users of these outfits and, through the Analytic Hierarchy Process (AHP), sorts out the weightings of demands of the rescuers; the method of Quality Function Deployment (QFD) is adopted to consolidate and summarize the design factors of multifunction inflatable blanket for emergency rescue in terms of function and usage, and improve the design concepts. Result: The design concepts of multi-function inflatable blankets for emergency rescue are formulated on the basis of AHP/QFD. In addition, it is verified that the multi-function inflatable blanket for emergency rescue is obviously advanced in terms of portability, easy operation, multifunctionality, rescue efficiency, and humanistic care, through the fabrication of samples and tests. Conclusion: It is proved that the application patterns of AHP and QDF theoretical models can play active roles in improving the design of products for emergency rescue. The design practice proves that this design scheme to some extent is feasible. The design method discussed in this study can serve as the reference for the design of other types of products for emergency rescue to some extent.

Aluminium profiles have an increasing demand since its excellent physical properties. At the same time, the crosssection shape of the profile is becoming more and more complex, which makes the design of its extrusion die more difficult and time consuming. To fill this gap, this paper has proposed an intelligent design method based on capturing and reusing the design knowledge of the extrusion die. Specifically, a knowledge representation model is created to reveal the key design knowledge generated during the design process. Then, a multi-modal knowledge graph is built to systematically structure and store the design information and knowledge. Finally, a smart design system is developed for the application of knowledge capturing and rueing by semantic retrieval and intelligent recommendation. Through applying and evaluating in a leading aluminium profile enterprise, this method has shown its ability in helping engineering designer in design knowledge reuse and improving design efficiency.

The global outbreak of Covid-19 causes a large number of casualties. Due to the suddenness and infectivity of virus, the bereaved were unable to say goodbye to the deceased. This leads to a series of psychological disorders of grief. In response to the above problems, a virtual reality(VR) grief farewell scene is designed to help bereaved children bid farewell to their deceased relatives and regain their confidence in life. To verify the effectiveness of the scene, 30 respondents were recruited by us to conduct a user experience survey. To a certain extent, the results show that grief empathy can be aroused by our design and the VR scene can assist subjects to vent their emotions and complete grief cognitive education. In terms of user experience, the scene fluency and refinement are basically intact. In future work, the VR grief scene design will be improved and more in-depth research on grief counseling will be conducted.

With the development of the technology era, people's demand for the experience of things is increasing. As a multi-edge discipline, digital media art is creative, interactive and aesthetic in practice, and can bring people an immersive emotional experience. This article designs and displays the cultural image of colleges and universities through digital media art, and constantly explores more displayable styles and creative forms in the design of college cultural images. Using the immersive experience to effectively express the campus cultural image through the scene design, it provides new ideas for the dissemination of the university image.

It is very important to study the filling gas ratio and the real-time change of halogen gas ratio for the performance optimization of GM tube. It is very time - and labor-saving to use the simulating software, but there are few reports about it. In this paper, based on Garfield++ software, the type and proportion of gas filling in GJ4105 GM tube are simulated, mainly for binary (neon + halogen) and ternary (neon + argon + halogen) gas filling. According to the dead time and signal amplitude of charge signal output by GM tube, the relationship between filling gas ratio and performance effect was studied, which provided technical basis for optimal design and subsequent improvement of halogen GM counter performance.

Under the guidance of information ecology theory, this paper discusses the design of the human-machine interaction framework of smart home terminals from the perspective of information ecology, and puts forward some problems existing in the current smart home niche. Based on the existing smart home terminals and the human-machine interaction theory, the smart home ecological industrial chain is established as a new system design model. The new humancomputer interaction system model and smart home ecological industrial chain will deeply interact with users, manufacturers, and service providers through smart home products, forming a closed industrial loop, filling the vacancy of the smart home niche, and providing inspiration for the design of smart home terminals.

AR technology has the potential to revolutionize the way we live or learn. The blackboard-style product structure teaching mode can no longer meet the real-time and immersive learning needs of students. Therefore, based on mobile AR technology, a platform for product practice teaching is designed. Combined with experiments, it focuses on the design of platform gesture, voice, and eye movement interaction. The platform usability test results show that it can well meet the learning needs and operating habits of students, increase the fun of learning, and provide new ideas for educational reform.

Aiming at handling the problems of long transmission lines and large floor area of the transmission device of double-shaft mixers, the Theory of Inventive Problem Solving (TRIZ) was adopted to analyze technology contradictions identified during the improvement of the transmission device and determine the factors to be improved and unfavorable factors to be weakened or eliminated. Then a double-output shaft worm gear reducer with a compact structure and small floor area was designed under the enlightenment of the inventive principle. Force analysis and strength checks were conducted for the worm shaft, and the mechanical simulation was carried out for the worm using ANSYS Workbench software. The results showed that the strength and stiffness of the worm shaft met the design requirements.

With the development of society, the country has put forward strict requirements for the economy and emission of motorcycles. As the most basic test platform, the small engine platform is of great significance to the improvement of the performance of small engines with relatively backward technical level. According to the small engine parameters, size, performance and other aspects of the requirements, to complete the overall design and layout of the small engine bench, clear the small engine bench simple, easy to install, stability, easy late maintenance and other target functions. According to these target functions to select materials, to construct a simple and safe small engine frame. Compared with the traditional engine platform function is single, the overall design is relatively large, bulky and other problems, the small engine test bench scheme is designed.

For a nuclear power plant cold source intake interceptor gear style and collection material cleaning needs, parameters and working parameters, the design and selection of various parts of the suction equipment combined with the current situation of the interceptor net, analyzed the feasibility of the interceptor net trash pumping program, and proposed a net interceptor pumping equipment design program with impurity pump as the main body. Through the analysis of environmental parts were completed, and the equipment manufacturing and testing conditions were available, which can improve the efficiency of garbage salvage in normal times and during marine life outbreak, and reduce the risk of on-site operation.

In order to guide the formula design of the Jiangsu China Tobacco brand, for the tobacco raw materials in the warehouse, 197 grades of tobacco samples that can represent the efficacy positioning of each cigarette formula are selected according to the features and efficacy positioning of different cigarette brands, and the cigarettes brand formula designs are guided by the feature display index. To understand the feature applicability of tobacco raw materials to cigarette efficacy categories according to the feature display index, the quantity ratio between different tobacco raw materials can be quickly carried out, thereby improving the efficiency of formula design and the quality of finished products. The results show that if the feature display index of a certain category of formula efficacy is greater than 100, it indicates that the display level is high; if the index is smaller than 100, it means that the display level is low.

With the continuous development of Chinese society and the gradual increase of the aging population, the product development of elderly products has also received more and more attention. In order to respond to the development of the market and the needs of users, product designers need to continuously optimize products according to the real needs of users, so as to improve the competitiveness of products. Based on the KANO model, this research has an in-depth understanding of the inner needs of users, combined with the research method of Kansei Engineering, extracts relevant design elements, and finally establishes the connection between user needs and product samples, so as to provide designers with help in product design and development. People-oriented design concept is the core concept of product design. Through the analysis and summary of this study, it is worth looking forward to improving the humanization of blood glucose meter products.

Positive systems are one class of systems with special properties that are prevalent in both macroscopic and microscopic domains. Stochastic positive systems have important theoretical value and wide application value in biology, economics and other fields. Therefore, the research on stochastic positive systems is of great significance. The paper investigates the stability with probability 1(w.p.1.) and stabilization designs of the discrete linear stochastic positive systems. Firstly, based on the comparison principle, the conditions for discrete linear stochastic positive systems with state probability constraints to be the stability with probability one is proposed. Secondly, based on this, the problem of stabilization for discrete linear stochastic controlled positive system is considered. Some sufficient conditions are proposed for constructing stable w.p.1 feedback controller. Finally, we verified the validity of the conclusions with numerical arithmetic examples based on linear programming theory.

Aims The software interface design is changing and updating and has spawned a large number of interface icon design requirements. Using deep learning and generative adversarial network(GAN) to assist icon design can simplify the process of icon design and enrich the creativity of icon conception. Methods The Python tool was used to capture nearly 2000 icon data and perform a simple filtering. After the steps of icon preprocessing, data transformation, structure design of generating network and discriminant network, definition of loss function, model recording and saving, a generation adversarial network suitable for icon design is successfully constructed. Results In this study, icon generation is well explored by combining deep learning and generative adversarial networks. Although the change of function loss and the output icon quality are not ideal in the preliminary practice, higher quality and new icons are obtained by improving the icon recognition algorithm, and the function loss value is reduced. The research has certain guiding significance for more personalized, higher quality and efficient icon generation and design in the future. Conclusion It is feasible to apply deep learning and GAN into icon design. The newly generated icons have certain artistic expression and instructional intention, and can be used as ideas and creativity in icon design process.

VR has been a hot research topic in recent years. Although VR on smartphone has been used for a long time, the experience of using it is far from enough. In scenery-based VR videos, the smartphone terminal, as the most extensive terminal, has a larger scope of promotion, and it is urgent to optimize the user experience. This paper studies the VR video with smartphone as the terminal and scenery as the video content from the perspective of interaction, and explores the interaction characteristics and interaction factors that affect the user experience.

In the trade of the ancient Chinese Silk Road, glass was a valuable material evidence of the early trade. However, the glass has experienced historical sand for a long time, and its surface is easily affected by the buried environment and weathering. In this paper, we will study the change of elements during weathering and identify and classify the glass based on the corresponding characteristics. Firstly, the component data of the ancient glass relics are processed uniformly, and the correlation analysis is conducted, and the model of the Logistics equation and the SVM subclassification model are trained and optimized with large lot of data to adjust to the best classification model, and the classification results. Through case analysis and comparative study of model results, it is shown that the classification model using potassium, silicon, lead and barium content as indicators has good applicability, accuracy and simplicity, which provides new ideas for the identification and analysis of the chemical composition of glass cultural relics.

A model was established to evaluate users' experiences when visiting the virtual reality museum. By using the shadowing survey method (N=6), context mapping (N=10), and expert meetings (N=4), 19 comprehensive experiences and 24 interactions were obtained, and an initial questionnaire model was obtained after summarizing and organizing them with correspondence analysis and cluster analysis. After the experimental evaluation and factor analysis (N=75), a questionnaire containing 14 relevant questions was derived to measure the user's experience in the virtual reality museum. With the development of information networks and virtual reality technology, virtual museums have been created to enable people to break the time and space constraints. While previous research on the museum experience has focused on real museums and virtual reality museum environments and technologies, this questionnaire brings a user perspective to the evaluation that can be used to promote and apply virtual reality museums in the future. The developed questionnaire allows for a comprehensive assessment that will help inform the future VR museum experience.

Data integration technology is of great significance to all kinds of analysis and decision-making systems based on largescale data. The difficulty of data integration lies in the diversity of data sources, data types and data storage methods, which makes it difficult to integrate data in real time. To solve the above problems, this paper designs a scheme, which uses the real-time stream processing characteristics of Apache Flink to realize the real-time integration of multi-source heterogeneous data through log capture and user-defined data processing. Through this scheme, multi-source heterogeneous data can be integrated into one place to provide data support for all kinds of big data analysis and decision-making.

Robots require both high precision localization and mapping solutions in unknown environments. Simultaneous localization and mapping is well matched to the requirement. In this article, the process of transforming SLAM problem into a state estimation problem is presented. Some implementations of SLAM are introduced. EKF-SLAM is easy to implement, but unable to build large maps, and unable to deal with nonlinear errors well. Fast-SLAM introduces particle filtering on the basis of EKF-SLAM, which solves nonlinear errors and increases the computation speed significantly. Graph-SLAM is based on graph optimization, able to build large maps and has good real-time performance. Visual SLAM and semantic SLAM are developing rapidly and will be the key development of SLAM in the future. The characteristics of each method are corroborated in the experimental results.

In order to study the movement of a hexapod robot in a low gravity environment, the feasibility of using the hexapod structure as the movement mode of the robot in the earth's outer environment is verified. In this paper, the structure and motion characteristics of the robot are constructed according to the hexapod beetle. The stability of the hexapod robot's typical triangular gait is analyzed using the barycenter projection method and introducing the concept of minimum stability margin. The simulation model of the hexapod robot is established using the mechanical system dynamics simulation software, and the motion simulation of the robot is analyzed. Through the analysis of the kinematics and dynamics parameters, the stability of the triangular gait, the rationality of the hexapod structure, and the feasibility of the robot’s movement in a low gravity environment are verified. It lays a foundation for future research on robot control and other gait stability and provides a theoretical basis for developing hexapod and other types of multi-legged robots.

With the development of big data, artificial intelligence, and wearable technology, the data generated by learners in the learning process can be fully recorded and stored. Using these data to study the characteristics of different types of learners and provide them with precise and personalized learning guidance is an urgent problem to be solved. This paper proposes a learner portrait construction method based on learning analysis technology. Firstly, the case frame relationship with "learning" as the central verb is established by using case grammar. Secondly, constructing a "learning-learner" feature layer with 6 sub-layers through qualitative analysis, and organizing the primary data of learners into 6 kinds of integrated data with specific meanings. Then, two methods of classifying learner types, cluster analysis and standard establishment, are proposed. Finally, giving an example of learner portrait construction.

The designing of network performances indexes are critical to respond to the factors that assess network performances. In this study, a novel assessment method for designing network performance indexes based on combined trapezoidal and intuitionistic fuzzy information axiom. First, the network performances indexes are designed with the customer-expected functional requirement (FR) and the existing FR as the design range and the system range respectively. Then the information content of qualitative and quantitative indexes, which facilitates the selection of the best alternative with minimum information content, are calculated separately and standardized. To solve this issue of the integral characterised by fuzzy bounds, this study proposes a defuzzification method to convert fuzzy bounds into crisp numbers. Finally, the assessment is performed to demonstrate the effectiveness of the developed approach.

Current state-of-the-art change detection networks depend on siamese CNN backbones to extract powerful feature representations in multi dimensions. However, architectures of existing siamese network methods are bloated and timeconsuming because they simply use whole backbones mentioned above in siamese architecture and roughly fuse all the information. The important step of the task is to design an effective feature extractor and fusion module that can enhance the performance of the corresponding change maps. In this paper, we propose a novel framework named Siamese Partial Change Network (SPCN) for fast and accurate image change segmentation. On the one hand, we modify backbones to extract change features partially in small resolution which are lightweight and more suitable for change detection tasks. On the other hand, we explore an alternative to integrate and refine change features for better results. Extensive experiments on CDnet 2014 dataset show that our proposed model not only outperforms previous state-of-the-art change detection methods, but also runs much faster than most existing models.

No-reference image quality assessment is the prediction of image quality without the original image and has a wide range of applications in fields such as image recognition. The wide content of images and the diversity of distortions acquired by mobile devices, make it a challenge to evaluate the quality of such truly distorted images captured by mobile devices. Therefore, in this paper, we propose to use global and local features of images to deal with the diversity of image distortions, using an attention module to extract global features to improve the characterization ability of global features. The global semantic features are used as the input of an adaptive weight and bias generation network, which can generate the weights and biases of the final quality prediction network based on the image content. In this paper, experiments are conducted on three real distortion databases, KonIQ-10k, BID, and LIVEC, and the results show that the proposed method outperforms existing no-reference image quality assessment methods. Also, experiments are conducted on two synthetic distortion databases LIVE and TID2013, and the results show that the method also achieves competitive results in evaluating synthetic distorted images.

Existing map matching methods determine the best driving paths for floating cars according to the constraints of the roads in the trajectory while ignoring the influence of traffic rules, which reduces the road path matching accuracy and increases road mismatch in complex situations with large road networks. In view of this phenomenon, this paper proposes a floating car trajectory map matching method based on traffic rule constraints that uses floating car retrograde and speeding behaviors and floating car positioning errors to match trajectory points. Four floating car trajectory sequences in Futian District, Shenzhen City, with a total length of 44.437 km are selected for verification, yielding 296 trajectory points. The experimental results show that the average matching accuracy of the proposed method in a complex road environment is 92.91%, which is approximately 7% higher than that of traditional methods. The proposed method meets user position accuracy requirements in floating car trajectory big data analyses and applications.

Motion Planning is an essential module for quadrotors, especially for automated navigation tasks. This paper introduces a search-based motion planning method for the quadrotor. The method can generate minimum time and smooth trajectory for quadrotors in a cluttered environment. The method explores the space with short-duration motion primitives based on obstacle information and dynamic constraints and generates an induced finite lattice discretization map. Then, the searchbased algorithm traverses through the map and returns a preferred trajectory. With the help of the heuristic function, the trajectory planning only takes a short time and is available for fast replanning during the flight. The performance of the proposed approach is evaluated using simulation and comparison with other state-of-art trajectory generation approaches.

Recommender systems serves as an important technical asset in many modern companies. With the increasing demand for higher precision of the technology, more and more research and investment has been allocated to the field. One important sub-field of recommender systems that has been stagnating is context-aware recommender systems. Due to the difficulty of collecting input dataset, the amount of research on context-aware recommender systems is much less than other sub-fields of recommender systems. In this paper, we propose a new algorithm named PowerMat to tackle the context-aware recommendation problem. We build our theory on matrix factorization and Zipf’s law, and also more recent research work such as DotMat. We prove by experiments that our method achieves superior results to the classic matrix factorization algorithm and other context-aware recommender systems such as MovieMat+. In addition, by theoretical analysis, we show that our algorithm solves the cold-start problem for context-aware recommendation.

In recent years, the world has set off a wave of manufacturing upgrading, and various enterprises have begun to transform into intelligent manufacturing. Through the analysis of the enterprise production process, to improve the efficiency and production efficiency of enterprises. In the process of research, the design algorithm of enterprise production scheduling process is one of the core of enterprise intelligent manufacturing research, it is also the focus of this paper. This paper first introduces the common production scheduling problems and the research status at home and abroad, then summarizes the scheduling rules and algorithms classification, and additionally explains the necessity of research on production scheduling problems, and then analyzes the advantages and disadvantages of related algorithms, and introduces the common methods to solve practical problems. Finally, the advantages of production scheduling algorithm in the process of enterprise transformation are described, and the future research plan is also explained.

Cardiovascular Diseases (CVDs) have become increasingly crucial in recent years and have been regarded as the leading cause of death worldwide. Although it is necessary to detect and treat CVDs in their early stages, only 67% of heart diseases could be predicted by medical professionals. Motivated by recent advances in Graph Neural Networks (GNNs) that have ramified in a variety of industries, in this paper, we utilize three novel GNN models, including Graph Convolutional Network (GCN), GraphSAGE, and Graph Attention Network (GAT), to conduct the heart disease prediction task, comprised of two stages: table-to-graph transformation and Graph Neural Networks prediction. Experimental results show significant improvements with the utilization of GNNs compared with three novel Machine Learning models: Logistic Regression (LR), Naïve Bayes (NB), and Multi-Layer Perceptron (MLP), and GAT performs optimally among the three GNN models. To the best of our knowledge, this is the first work that predicts heart disease using GNNs.

Multiple object tracking is a field with extensive practical applications. The problem of multiple object tracking can be broken down to tracking and association. The algorithms of common tracking and association algorithms are described in this paper and their relative characteristics described. Two different association algorithms, Global Nearest Neighbor and Multiple Hypothesis Tracking are compared using a MATLAB simulation run on different test footage. The characteristic features with regards to track creation and assignment accuracy of the two are shown by their performance in both cases.

Automatic modulation classification (AMC) recognizes modulation patterns based on features extracted from received signals. Benefiting from its powerful feature extraction capabilities, deep learning is widely and successfully applied for AMC. However, existing AMC mainly based on supervised deep learning, which is highly dependent on labelled data and has difficulty constructing large-scale and well-labelled datasets, still faces challenges. To address this challenge, we propose an unsupervised learning architecture for AMC (USLAMC) by uniting feature extraction and deep subspace clustering (DSC). Feature extraction can utilize multi-level information to obtain the original dataset features. DSC can generate self-representation relationships for features, and use spectral clustering realizes unsupervised subspace feature clustering.

In the case where prior knowledge such as frequency, bandwidth, and modulation mode is unknown, the RF(radio frequency) signal from the receiver must be demodulated to a bitstream. In this paper, unsupervised clustering of logical channels such as communication system signals, service synchronization or physical layer signaling using deep learning methods is analyzed. After analysis, it is found that the current channel clustering mainly faces two problems: one is the need to manually annotate the key features of the signal dock; Second, different representations of signals, such as square spectrum and time-frequency graph, contain different feature information. At present, most of them are only analyzed for specific signal representations. Therefore, we propose to use the mixed-domain attention mechanism to automatically locate the areas that need to be focused on instead of manual participation. At the same time, use product layer for feature fusion of different signal representations (Identical/orthogonal format, Amplitude/Phase format, Time and Frequency diagram) to improve the recognition accuracy, which is a feasible direction for future research.

Chemical industry is of importance in China both strategically and economically, which plays a key role, whereas chemical products are usually harmful. In light of this, their transportation must be ensured to be safe. In this work, the leakage of the emergency shut-off valve and the sealing surface of a tank body for anhydrous hydrogen fluoride is investigated, through which the reason causing the leakage is analyzed systematically in order to make a reasonable emergency response, aiming to minimize personnel and property losses. The results of this study can provide referred emergency response for the equipment of the same type, which is helpful in reducing the leakage probability in the daily usage and management.

At present, the design of the vehicle guided by the vertical vibration characteristics of the human body is mainly based on the ISO5982-2001 standard, but the standard is mainly based on the body structure characteristics of Europeans and Americans, and is not necessarily suitable for the physique of Chinese. In order to better guide the design of Chinese vehicles, this paper built a test bench for the study of vertical vibration characteristics of the human body and conducted relevant experiments on 20 volunteers. The relevant data of 20 volunteers such as apparent mass, drive point mechanical impedance and seat-head transfer function were obtained through the experiment, and compared with the international standard ISO5982-2001. It is found that the vertical vibration characteristics of the human body under the standard sitting conditions of the Chinese are quite different from the ISO5982-2001 standard, so the relevant indicators of the ISO5982-2001 standard on the vertical vibration characteristics of the human body under the standard sitting conditions are not applicable to the design of guiding Chinese vehicles.

With the development of automobile market, new energy vehicles have reached the peak trend. As a pure electric vehicle corner overtaking China, the development of electric vehicles should not be underestimated. According to the investigation, pure electric vehicles generally use fixed gear ratio reducer, which is far from the performance of traditional multi-gear transmission. Pure electric vehicles are driven by motors, and the shift schedule of transmission vehicles cannot be rigidly nested on pure electric vehicles. Therefore, it is of great significance to study the multi-gear shift law of pure electric vehicles.