Back-end Module Research Articles

RS-SLAM: real time semantic slam with driverless car using LiDAR-camera-IMU sensing

Abstract Accurate and robust Simultaneous Localization and Mapping (SLAM) technology is a critical component of driverless cars, and semantic information plays a vital role in their analysis and understanding of the scene. In the actual scene, the moving object will produce the shadow phenomenon in the mapping process. The positioning accuracy and mapping effect will be affected. Therefore, we propose a semantic SLAM framework combining LiDAR, IMU, and camera, which includes a semantic fusion front-end odometry module and a closed-loop back-end optimization module based on semantic information. An improved image semantic segmentation algorithm based on Deeplabv3+ is designed to enhance the performance of the image semantic segmentation model by replacing the backbone network and introducing an attention mechanism to ensure the accuracy of point cloud segmentation. Dynamic objects are detected and eliminated by calculating the similarity score of semantic labels. A loop closure detection method based on semantic information is proposed to detect key semantic features and use threshold range detection and point cloud re-matching to establish the correct loop closure detection, and finally reduce the global cumulative error and improve the global trajectory accuracy using graph optimization to ultimately obtain the global motion trajectory and realize the construction of 3D semantic maps. We evaluated it on the KITTI dataset and collected a dataset for evaluation by ourselves, which includes four different sequences. The results show that the proposed framework has good positioning accuracy and mapping effect in large-scale urban road environments.

Design of a JSP based Mobile Phone Repair Reservation System

With the development of the times, mobile phones are becoming more and more popular. The increase in the number of mobile phones will lead to an increase in the number of mobile phone repairs. Developing a mobile phone repair appointment system can be beneficial for mobile phone repairs. The mobile phone repair reservation system adopts the B/S mode, using Eclipse software as the system development tool and Java programming language for system development. The system database adopts SQL Server database. In system development, JSP and Servlet technologies are used to help users repair their phones more conveniently, while also benefiting maintenance personnel. The mobile phone repair reservation system designed in this paper consists of a front-end user module and a back-end administrator module. The front-end module implements functions such as registering and logging in for front-end maintenance personnel, modifying personal information, publishing, viewing, modifying, and deleting appointments for maintenance personnel, registering and logging in for front-end users, modifying personal information, making appointments, viewing appointments, and canceling appointments. The backend module enables administrators to modify personal information, view and delete personal information of users and maintenance personnel, view and delete appointment information, and other functions.

Sustainable scheduling of TFT-LCD cell production: A hybrid dispatching rule and two-phase genetic algorithm

TFT-LCD manufacturing process involves Array, Color Filter (CF), Cell, and Module. The Cell process is pivotal in the TFT-LCD production supply chain; it bridges the front-end (Array and CF) and back-end (Module), maintaining a balance of production capacities. Supply from the front-end is transmitted to the back-end, ensuring prompt responsiveness to customer demands. As awareness of environmental conservation grows, implementing sustainable development in the TFT-LCD manufacturing industry has become crucial. Specifically, reducing the machine setup frequency in TFT-LCD scheduling decreases energy consumption during machine operation and wait times and prolongs equipment lifespan by minimizing wear from repeated shutdowns. To tackle the intricate scheduling challenges within TFT-LCD production, this study proposes an intelligent, sustainable scheduling approach that employs a rolling dispatching and line balancing mechanism and a two-phase genetic algorithm (HDTPGA) for Cell scheduling. The goal is to minimize makespan considering product diversity, production constraints, work-in-process balancing, and machine setup frequency. This holistic approach promotes energy-efficient and sustainable manufacturing practices. According to the empirical study, the proposed HDTPGA reduces the makespan by approximately 10% compared to the baseline. Additionally, it reduces the average daily setup frequency of bottleneck stages (ODF and CCT) in the Cell process by approximately 68.25% and 66.25%, respectively. The main contribution of this paper is to propose an intelligent scheduling algorithm to optimize Cell production scheduling in practice. The HDTPGA achieves equilibrium between front-end and back-end production, utilizing an efficient line change mechanism to reduce energy consumption. The enhancement in efficiency not only restricts material waste but also reduces the environmental impact, enabling manufacturers to adopt green production practices and promoting a transition of the entire panel industry toward sustainability.

Stock Quantitative Intelligent Investment Model Based on Machine Learning Algorithms

In order to cope with sudden changes in market style, the author designed and implemented an investment system. In response to the inability to cope with sudden changes in market style, the author proposed an improved scheme when using a classifier for training, using a hidden Markov model to select time points in the same market state as the training samples for the classifier. The results of backtesting and real market testing show that the improved investment system can capture changes in market style, and during the testing period, the test results obtained higher returns compared to the pre improved classifier. At present, the investment system implemented by the author has become the core backend module of the officially launched intelligent investment advisory product Zhiyu Liangtou APP, providing strong investment decision-making suggestions for small and medium-sized investors. In the future, we can provide backend support for more user end products.

Development of the Backend Module of the Agroecological Land Suitability Assessment System for Garlic

Development of the Backend Module of the Agroecological Land Suitability Assessment System for Garlic

A comprehensive overview of core modules in visual SLAM framework

Visual Simultaneous Localization and Mapping (VSLAM) technology has become a key technology in autonomous driving and robot navigation. Relying on camera sensors, VSLAM can provide a richer and more precise perception means, and its advancement has accelerated in recent years. However, current review studies are often limited to in-depth analysis of a specific module and lack a comprehensive review of the entire VSLAM framework. The VSLAM system consists of five core components: (1) The camera sensor module is responsible for capturing visual information about the surrounding environment. (2) The front-end module uses image data to roughly estimate the camera’s position and orientation. (3) The back-end module optimizes and processes the pose information estimated by the front-end. (4) The loop detection module is used to correct accumulated errors in the system. (5) The mapping module is responsible for generating environmental maps. This review provides a systematic and comprehensive analysis of the SLAM framework by taking the core components of VSLAM as the entry point. Deep learning brings new development opportunities for VSLAM, but it still needs to solve the problems of data dependence, cost and real-time in practical application. We deeply explore the challenges of combining VSLAM with deep learning and feasible solutions. This review provides a valuable reference for the development of VSLAM. This will help push VSLAM technology to become smarter and more efficient. Thus, it can better meet the needs of future intelligent autonomous systems in multiple fields.

Evaluation of the astrometric capability of Multi-Channel Imager and simulated calculation of proper motion of M31

Introduction: The Multi-Channel Imager (MCI), one of the back-end modules of the future China Space Station Telescope (CSST), is designed for high-precision spacebased astronomical observations. This paper evaluates the astrometric capability of the MCI based on simulated observational images and Gaia data: the M31 galaxy is selected as a representative case to validate the astrometric capability by calculating the proper motions (PMs) of the M31 member stars.Method: We analyze the stellar centroids of the simulated images in the R, I and G bands, positional uncertainty of 2.5 mas for brighter foreground reference stars from the Gaia DR3 catalog and of 7.5 mas for the fainter M31 member stars, are adopted respectively. The theoretical PMs are generated from the adopted velocity field model, rotation curve, and stellar surface density profile. And the simulated observed PMs are generated from the aforementioned position uncertainties and theoretical PMs.Result: We conclude that the precision of the MCI derived PMs strongly depends on the number of astrometric epochs per year. Specifically, uncertainty of 10 μas/yr is achievable with 10 epochs per year, and of 5 μas/yr with 50 epochs ignoring possible systematic effects. And symmetrically distributed observed fields yield better M31 kinematic parameters.Discussion: Unknown systematic errors, space environment effects on detectors, dithering strategies, and observation schedules can affect the PMs of M31, the above issues need further analysis and validation in future work.

A 0.67 μV-IIRN super-T Ω-Z IN 17.5 μW/Ch Active Electrode With In-Channel Boosted CMRR for Distributed EEG Monitoring.

We present the design, development, and experimental characterization of an active electrode (AE) IC for wearable ambulatory EEG recording. The proposed architecture features in-AE double common-mode (CM) rejection, making the recording's CMRR independent of typically-significant AE-to-AE gain variations. Thanks to being DC coupled and needless of chopper stabilization for flicker noise suppression, the architecture yields a super-T Ω input impedance. Such a large input impedance makes the AE's CMRR practically immune to electrode-skin interface impedance variations across different recording channels, a critical feature for dry-electrode ambulatory systems. Signal quantization and serialization are also performed in-AE, which enables a distributed system in which all AEs use a single data bus for data/command communication to the backend module, thus significantly improving the system's scalability. Additionally, the presented AE hosts auxiliary modules for (i) detection of an unstable electrode-skin connection through continuous interface impedance monitoring, (ii) dynamic measurement and adjustment of input DC level, and (iii) a CM feedback loop for further CMRR enhancement. The article also presents the development of printed (extrusion) tattoo electrodes and their experimental characterization results with the proposed AE architecture. Besides bio-compatibility, low-cost, pattern flexibility, and quick fabrication process, the printed electrodes offer a very stable electrode-skin connection, conform to scalp shape, and exhibit consistent performance under various bending curvatures. Analog circuit blocks of the presented AE architecture are designed and fabricated using a standard 180 nm CMOS technology, and the [Formula: see text] IC is integrated with off-chip low-power digital modules on a PCB to form the AE. Our measurement results show a CMRR of 82.2 dB (at 60 Hz), amplification voltage gain of 52.8 dB, a bandwidth of 0.2-400 Hz, ±500 mV input DC offset tolerance, An input impedance [Formula: see text], and 0.67 μV RMS integrated input referred noise (0.5-100 Hz), while consuming 17.5 μW per channel. All auxiliary modules are tested experimentally, and the entire system is validated in-vivo, for both ECG and EEG recording.

A cloud infrastructure for enabling provision of the Material Testing as a Service (MTaaS)

This paper focuses on a ‘cloud infrastructure’ (MTaaS) that has been developed to support and coordinate integrated services and supply chain management. This will enable the seamless operation of a dynamic and adaptive system providing customer-centred material testing services. The cloud system is designed not only for decision making regarding the evaluation of the situations brought by industry and translating them into upscaling, characterisation and service workflows, but also for promoting existing products, improving facility usage and providing expertise consultancy. The development of MTaaS involves specifying the cloud infrastructure requirement, defining the back-end and front-end system architecture and modules, and outlining the operation fields of each module along with associated terminology, parameters and contents. This includes handling material testing service registration, verification, marketing, composition, preparation, offering, processing, monitoring as well as customer inquiries, monitoring and feedback. These efforts make MTaaS integrate product, service and customer-related information into a single platform, encompassing necessary operation related data, customer related data, inventory data and supply chain data.

Intelligent Design and Application of Traditional Cultural and Creative Products Based on Digital Art Elements

Abstract Cultural and creative industry is one of the main fields of digital art element design, and this paper utilizes digital art design for visualization programming of cultural and creative works. Through the three elements of 3D model, registration method and animation composition, the intelligent design process of AR cultural and creative products is defined, and the editing process of AR cultural and creative products is sorted out on this basis. Construct the model of cultural and creative products containing digital art elements, solve the problem of inconsistency between the coordinates and the model network appearing in the modeling, and then carry out the back-end design processing for AR cultural and creative products editing, which constitutes the whole back-end module through real-time methods and multiple trigger events. Through the method of empirical analysis, the feasibility of adding digital art elements to cultural and creative products, the consumers’ willingness to purchase cultural and creative products, and the innovation performance of cultural and creative products are analyzed. In the test of the moderating effect of the intelligent design of digital art elements on the cultural and creative products, the performance of the cultural and creative products improves synchronously with the embedding of the digital art elements, and the mean value finally stays at 0.375, and all the straight lines are above zero.

Multiagent Soft Actor-Critic Based Hybrid Motion Planner for Mobile Robots.

In this article, a novel hybrid multirobot motion planner that can be applied under no explicit communication and local observable conditions is presented. The planner is model-free and can realize the end-to-end mapping of multirobot state and observation information to final smooth and continuous trajectories. The planner is a front-end and back-end separated architecture. The design of the front-end collaborative waypoints searching module is based on the multiagent soft actor-critic (MASAC) algorithm under the centralized training with decentralized execution (CTDE) diagram. The design of the back-end trajectory optimization module is based on the minimal snap method with safety zone constraints. This module can output the final dynamic-feasible and executable trajectories. Finally, multigroup experimental results verify the effectiveness of the proposed motion planner.

A review of piezoelectric-electromagnetic hybrid energy harvesters for different applications.

Social progress is inseparable from the utilization of energy, signals of extreme consumption of fossil energy and energy crisis appear frequently around the world. Human beings are paying more and more attention to new technologies and the sustainable development of energy collection and conversion. The emergence of piezoelectric, electromagnetic, electrostatic, and triboelectric mechanisms provides a variety of effective methods for new environmental energy collection and conversion technologies. Among them, the piezoelectric-electromagnetic hybrid energy harvester (P-EHEH) has been widely studied due to its high output power, simple structure, and easy miniaturization. Continuous progress has been made in the research of P-EHEH through theoretical exploration, structural optimization, and performance improvement. This Review focuses on the review of P-EHEH at the application level. A detailed introduction summarizes the research status of P-EHEH applied to human body devices, monitoring sensors, and power supply devices, as well as the development status of back-end electronic modules and interface circuits. The future challenges and development prospects of P-EHEH are anticipated.

ODLC_SAM: a novel LiDAR SLAM system towards open-air environments with loop closure

PurposeThe light detection and ranging sensor has been widely deployed in the area of simultaneous localization and mapping (SLAM) for its remarkable accuracy, but obvious drift phenomenon and large accumulated error are inevitable when using SLAM. The purpose of this study is to alleviate the accumulated error and drift phenomenon in the process of mapping.Design/methodology/approachA novel light detection and ranging SLAM system is introduced based on Normal Distributions Transform and dynamic Scan Context with switch. The pose-graph optimization is used as back-end optimization module. The loop closure detection is only operated in the scenario, while the path satisfies conditions of loop-closed.FindingsThe proposed algorithm exhibits competitiveness compared with current approaches in terms of the accumulated error and drift distance. Further, supplementary to the place recognition process that is usually performed for loop detection, the authors introduce a novel dynamic constraint that takes into account the change in the direction of the robot throughout the total path trajectory between corresponding frames, which contributes to avoiding potential misidentifications and improving the efficiency.Originality/valueThe proposed system is based on Normal Distributions Transform and dynamic Scan Context with switch. The pose-graph optimization is used as back-end optimization module. The loop closure detection is only operated in the scenario, while the path satisfies condition of loop-closed.

Web-Based Application Based on Human-in-the-Loop Deep Learning for Deidentifying Free-Text Data in Electronic Medical Records: Development and Usability Study.

The narrative free-text data in electronic medical records (EMRs) contain valuable clinical information for analysis and research to inform better patient care. However, the release of free text for secondary use is hindered by concerns surrounding personally identifiable information (PII), as protecting individuals' privacy is paramount. Therefore, it is necessary to deidentify free text to remove PII. Manual deidentification is a time-consuming and labor-intensive process. Numerous automated deidentification approaches and systems have been attempted to overcome this challenge over the past decade. We sought to develop an accurate, web-based system deidentifying free text (DEFT), which can be readily and easily adopted in real-world settings for deidentification of free text in EMRs. The system has several key features including a simple and task-focused web user interface, customized PII types, use of a state-of-the-art deep learning model for tagging PII from free text, preannotation by an interactive learning loop, rapid manual annotation with autosave, support for project management and team collaboration, user access control, and central data storage. DEFT comprises frontend and backend modules and communicates with central data storage through a filesystem path access. The frontend web user interface provides end users with a user-friendly workspace for managing and annotating free text. The backend module processes the requests from the frontend and performs relevant persistence operations. DEFT manages the deidentification workflow as a project, which can contain one or more data sets. Customized PII types and user access control can also be configured. The deep learning model is based on a Bidirectional Long Short-Term Memory-Conditional Random Field (BiLSTM-CRF) with RoBERTa as the word embedding layer. The interactive learning loop is further integrated into DEFT to speed up the deidentification process and increase its performance over time. DEFT has many advantages over existing deidentification systems in terms of its support for project management, user access control, data management, and an interactive learning process. Experimental results from DEFT on the 2014 i2b2 data set obtained the highest performance compared to 5 benchmark models in terms of microaverage strict entity-level recall and F1-scores of 0.9563 and 0.9627, respectively. In a real-world use case of deidentifying clinical notes, extracted from 1 referral hospital in Sydney, New South Wales, Australia, DEFT achieved a high microaverage strict entity-level F1-score of 0.9507 on a corpus of 600 annotated clinical notes. Moreover, the manual annotation process with preannotation demonstrated a 43% increase in work efficiency compared to the process without preannotation. DEFT is designed for health domain researchers and data custodians to easily deidentify free text in EMRs. DEFT supports an interactive learning loop and end users with minimal technical knowledge can perform the deidentification work with only a shallow learning curve.

Flexible Finger-Shaped Active Dry EEG Electrode With a Configurable Application-Specific Integrated Circuit and Embedded Leadoff Detection

Dealing with the discomfort and complex setting-up of conventional wet electrodes in the electroencephalography (EEG) monitoring for personal daily healthcare, a flexible finger-shaped active dry EEG electrode is developed and tested in this paper. The flexible finger-shaped dry electrodes, made of polydimethylsiloxane (PDMS) and elastic electrically conductive composite (ECC), penetrate through the hair to achieve stable contact with the scalp. To improve the quality of weak signal acquisition, a customized configurable application-specific integrated circuit (ASIC) is integrated using ECC based fabrication process, which amplifies the signal by 100 V/V, detects the lead-off states of electrodes, communicates with the back-end electronic module for configuration and status feedback. The amplifiers have >100 MΩ @ 50 Hz input impedance to match the large interface impedance of dry electrodes. An I2C slave controller with a dedicated wake-up oscillator is introduced for excellent configurability with a reduced number of wires and lower power consumption. Modulus of the proposed electrodes are similar as human tissue, ensuring remarkable user comfort. Besides, the process costs low and mold-casting is suitable for mass production. Over all, the easy-to-use active dry electrode could meet the requirements for daily EEG monitoring.

Design and Implementation of a C++Based Hotel Management System

In this paper, we use C++ as the development language and QT to design and develop a hotel management system for the current development of the hotel industry. The system adopts the B/S model to connect the object with the database, so that it has better flexibility and reusability. Its functional modules mainly include customer service module, front-end processing module, back-end management module, data storage module and other parts, which mainly realize the hotel customer's reservation, check-in and check-out and the management of the hotel internal staff and other functions. The development cost of the system is low and the operation is simple, which reduces the learning cost of the staff, improves the management efficiency of the hotel and facilitates the maintenance and management at a later stage.

ASL-SLAM: A LiDAR SLAM With Activity Semantics-Based Loop Closure

A practical back-end module with loop closure detection is very useful and important for a LiDAR simultaneous localization and mapping (SLAM) system to perform high-precision positioning and mapping tasks. However, most existing loop closure detection methods are based on images or point clouds, and these methods may produce errors when the structure or texture is similar. To overcome this problem, we propose a complete LiDAR SLAM system, including a front-end odometry module based on normal distribution transform (NDT)-LOAM and a back-end optimization module with loop closure based on activity semantics. Through the analysis and calculation of inertial measurement unit (IMU) data from SLAM platforms such as unmanned ground vehicles (UGVs), the activity semantics of turning and passing over a speed bump are detected based on the peak <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${z}$ </tex-math></inline-formula> -axis angular velocity and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${z}$ </tex-math></inline-formula> -axis acceleration, respectively. Then, according to this activity semantics information and its unique and definite attributes, we establish correct loop closure detection using rough geometric detection, activity semantics matching, and point cloud rematching for validation. Finally, graph optimization theory is utilized to reduce the global cumulative error, improve the global trajectory accuracy and map consistency, and obtain the final global motion trajectory and point cloud map. We collected a dataset for evaluation, which contains indoor data, outdoor data, and indoor–outdoor integration data, and we also evaluated our method on the KITTI dataset. The experimental results for different scenes show that the addition of activity semantics can effectively help loop closure detection and improve LiDAR SLAM system performance.

Development of a mobile application for forest and land fire patrol using Prototype method

A previous study has developed a mobile application for recording field data during Sumatra, Indonesia’s forest and land fires patrol. After users tested the usability, the user’s need for the system increased due to the new environmental characteristics in the forest and land fire patrol area. This study aims to develop a mobile application that is easy to use by the patrol team to support forest and land fire prevention patrols. The system development continues the mobile application proposed in the previous study. The system development method used is Prototyping, with stages starting from communication; quick design; rapid design modelling; prototyping; dissemination, delivery, and feedback. The application was built using the Ionic 3 framework using the REST API from the back-end module. The development was conducted in two iteration stages, namely the third and fourth iterations. The third and fourth iterations resulted in changes in the user interface, added auto-fill features, and downloaded reports. Applications developed in each iteration were tested using the Black-Box method. The test results from each iteration show that all functions in the application work properly and meet the user requirements.

CNTSeg: A multimodal deep-learning-based network for cranial nerves tract segmentation.

The segmentation of cranial nerves (CNs) tracts based on diffusion magnetic resonance imaging (dMRI) provides a valuable quantitative tool for the analysis of the morphology and course of individual CNs. Tractography-based approaches can describe and analyze the anatomical area of CNs by selecting the reference streamlines in combination with ROIs-based (regions-of-interests) or clustering-based. However, due to the slender structure of CNs and the complex anatomical environment, single-modality data based on dMRI cannot provide a complete and accurate description, resulting in low accuracy or even failure of current algorithms in performing individualized CNs segmentation. In this work, we propose a novel multimodal deep-learning-based multi-class network for automated cranial nerves tract segmentation without using tractography, ROI placement or clustering, called CNTSeg. Specifically, we introduced T1w images, fractional anisotropy (FA) images, and fiber orientation distribution function (fODF) peaks into the training data set, and design the back-end fusion module which uses the complementary information of the interphase feature fusion to improve the segmentation performance. CNTSeg has achieved the segmentation of 5 pairs of CNs (i.e. optic nerve CN II, oculomotor nerve CN III, trigeminal nerve CN V, and facial-vestibulocochlear nerve CN VII/VIII). Extensive comparisons and ablation experiments show promising results and are anatomically convincing even for difficult tracts. The code will be openly available athttps://github.com/IPIS-XieLei/CNTSeg.

Optimization of a Microwave Polarimeter for Astronomy with Optical Correlation and Detection

Cosmic Microwave Background (CMB) B-modes detection is the main focus of future CMB experiments because of the valuable information it contains, particularly to probe the physics of the very early universe. For this reason, we have developed an optimized polarimeter demonstrator sensitive to the 10-20 GHz band in which the signal received by each antenna is modulated into a Near Infrared (NIR) laser by a Mach-Zehnder modulator. Then, these modulated signals are optically correlated and detected using photonic back-end modules consisting of voltage-controlled phase shifters, a 90-degree optical hybrid, a pair of lenses, and an NIR camera. During laboratory tests, a 1/f-like noise signal related to the low phase stability of the demonstrator has been found experimentally. To solve this issue, we have developed a calibration method that allows us to remove this noise in an actual experiment, until obtaining the required accuracy level in the measurement of polarization.