Real-time Problem Research Articles

Low-Power 8T SRAM Compute-in-Memory Macro for Edge AI Processors

The traditional Von Neumann architecture creates bottlenecks due to data movement. The compute-in-memory (CIM) architecture performs computations within memory bit-cell arrays, enhancing computational performance. Edge devices utilizing artificial intelligence (AI) address real-time problems and have established themselves as groundbreaking technology. The 8T structure proposed in this paper has strengths over other existing structures in that it better withstands environmental changes within the SRAM and consumes lower power during memory operation. This structure minimizes reliance on complex ADCs, instead utilizing a simplified voltage differential approach for multiply-and-accumulate (MAC) operations, which enhances both power efficiency and stability. Based on these strengths, it can achieve higher battery efficiency in AI edge devices and improve system performance. The proposed integrated circuit was simulated in a 90 nm CMOS process and operated on a 1 V supply voltage.

Three six-degree-of-freedom dynamic descent convex optimization guidance methods

A key technology for soft landing of powered descent of aircraft is to solve the six-degree-of-freedom powered descent guidance problem in real time, which can be described as the fuel-saving trajectory optimization problem under multiple constraints. Three optimization models are established by selecting three independent variables: flight time, time substitution variable, and trajectory height. The original trajectory optimization problem is transformed into a solvable form of sequential convex optimization for iterative solution, forming three online guidance methods. The three guidance methods are compared in terms of convergence, real-time performance, optimality, and solution accuracy. The results show that all three guidance methods can solve the six-degree-of-freedom powered descent problem. The guidance method with flight time as the independent variable has the shortest calculation time and the least fuel consumption, but the powered descent flight time needs to be determined in advance. The guidance methods based on the other two types of independent variables can optimize the powered descent flight time, but they are all suboptimal solutions, and the calculation time is significantly increased. The solution accuracy of the three methods is similar under the same number of discrete points. If sequential convex optimization is used as the online guidance scheme for powered descent, how to determine the optimal flight time, approach the optimal solution for fuel, and further shorten the calculation time still needs to be further studied.

Optimization of multi-path approach flight sequencing based on rolling horizon control

Arrival flight sequencing optimization is an effective method to improve the landing efficiency of incoming flights and reduce flight delays. Based on this, with the goal of maximizing landing efficiency, combined with multi-runway and multi-route selection, and considering the actual waypoint restrictions, a mixed integer programming model for multi-path and multi-runway integrated arrival flight sequencing optimization is proposed. In order to solve the real-time problem of large-scale flight sequencing calculation, a multi-waypoint rolling horizon control algorithm is proposed. The terminal area of Guangzhou Baiyun International Airport is used as an example for verification, and the actual arrival flight data is used to carry out calculation experiments. In terms of wake turbulence safety interval, the RECAT-CN operation standard is used. The calculation results show that when the number of small-scale flights (23 flights) is small, the maximum landing time of the proposed model is 55 s earlier than the first-come-first-served method and 271 s earlier than the non-optimized one; when the number of large-scale flights (104 flights) is large, the solver alone cannot find a feasible solution within 3600 s, and the proposed algorithm finds a solution within 128.65 s. The proposed model and algorithm are effective and can be applied to actual flight scheduling optimization.

A Review of Research on Handwritten Chinese Character Recognition with Multi-Feature Fusion

This paper analyzes the progress of handwritten Chinese character recognition technology, from two perspectives: traditional recognition methods and deep learning-based recognition methods. Firstly, the complexity of Chinese character recognition is pointed out, including its numerous categories, complex structure, and the problem of similar characters, especially the variability of handwritten Chinese characters. Subsequently, recognition methods based on feature optimization, model optimization, and fusion techniques are highlighted. The fusion studies between feature optimization and model improvement are further explored, and these studies further enhance the recognition effect through complementary advantages. Finally, the article summarizes the current challenges of Chinese character recognition technology, including accuracy improvement, model complexity, and real-time problems, and looks forward to future research directions.

Efficient Network-based Fault Detection in Elevator Vibration Signals: A Weighted Fusion Approach of Displacement and Acceleration Data

This research addresses human vibration analysis, rule-based lift fault detection algorithm noise interference, and early fault diagnosis. A new network-based defect diagnosis system uses lift sensor acceleration and displacement data. Convolutional Neural Networks CNNs automatically extract features like motor current (average, maximum, lowest values during a recording session) from this fused data stream and sensor data. CNNs learn key features from pooled data to reduce noise and detect faults early. To build a baseline for normal operation, 132 vibration signal characteristics (RMS, peak-to-peak, mean, standard deviation), lift ID, timestamp, operating mode (Up/Down/Idle), and load condition (Empty/Low/Medium/High) data points are used. Regular operational data and 201 gearbox fault data points are collected. The problem affects motor behaviour via vibration signals, timestamps, lift data, and motor current values. Learning to distinguish this defect type from normal operational data allows the network-based technique identify it. Research shows that network-based strategies outperform traditional methods. It finds weaknesses better than before researches. This method automatically extracts features from the fused data stream to detect defects in real time. Displacement and acceleration data, motor current readings, and CNNs improve noise resistance and failure signal detection. Network-based lift failure detection is advanced in real time problem identification. Precision, early detection, noise resistance, and real-time defect categorization are its strengths. Study affects lift maintenance and find defects early to keep lifts functioning. Real-time defect classification eliminates laborious analysis, simplifying maintenance in which enhancements improve lift performance and reliability.

Inchworm quasi Monte Carlo for quantum impurities

The inchworm expansion is a promising approach to solving strongly correlated quantum impurity models due to its reduction of the sign problem in real and imaginary time. However, inchworm Monte Carlo is computationally expensive, converging as 1/N where N is the number of samples. We show that the imaginary-time integration is amenable to quasi Monte Carlo, with parametrically better 1/N convergence, by mapping the Sobol low-discrepancy sequence from the hypercube to the simplex with the so-called Root transform. This extends the applicability of the inchworm method to, e.g., multiorbital Anderson impurity models with off-diagonal hybridization, relevant for materials simulation, where continuous-time hybridization expansion Monte Carlo has a severe sign problem. Published by the American Physical Society 2024

Enhancing Manufacturing Quality Through Statistical Process Control

In the manufacturing industry, statistical process control, often known as SPC, is a strong instrument that can improve both the quality of the product and the efficiency of the process. It comprises the utilization of statistical approaches for the purpose of monitoring and controlling a process in order to guarantee that it functions effectively and generates products that are of a consistently high grade. The statistical process control (SPC) method enables producers to identify variances, trends, and irregularities that may have an impact on product quality. This is accomplished by collecting and analyzing data from the production process. It is possible for manufacturers to discover and rectify problems in real time with the assistance of statistical process control (SPC), which is accomplished through the utilization of control charts, process capability analysis, and other statistical approaches. This results in decreased waste, greater productivity, and increased customer satisfaction. This article examines the fundamentals of statistical process control (SPC), as well as its application in manufacturing environments and the advantages it offers in terms of improving both product quality and process performance.

Flexible control and trajectory planning of medical two-arm surgical robot.

This paper introduces the flexible control and trajectory planning medical two-arm surgical robots, and employs effective collision detection methods to ensure the safety and precision during tasks. Firstly, the DH method is employed to establish relative rotation matrices between coordinate systems, determining the relative relationships of each joint link. A neural network based on a multilayer perceptron is proposed to solve FKP problem in real time. Secondly, a universal interpolator based on Non-Uniform Rational B-Splines (NURBS) is developed, capable of handling any geometric shape to ensure smooth and flexible motion trajectories. Finally, we developed a generalized momentum observer to detect external collisions, eliminating the need for external sensors and thereby reducing mechanical complexity and cost. The experiments verify the effectiveness of the kinematics solution and trajectory planning, demonstrating that the improved momentum torque observer can significantly reduce system overshoot, enabling the two-arm surgical robot to perform precise and safe surgical tasks under algorithmic guidance.

A novel discrete zeroing neural network for online solving time-varying nonlinear optimization problems.

To reduce transportation time, a discrete zeroing neural network (DZNN) method is proposed to solve the shortest path planning problem with a single starting point and a single target point. The shortest path planning problem is reformulated as an optimization problem, and a discrete nonlinear function related to the energy function is established so that the lowest-energy state corresponds to the optimal path solution. Theoretical analyzes demonstrate that the discrete ZNN model (DZNNM) exhibits zero stability, effectiveness, and real-time performance in handling time-varying nonlinear optimization problems (TVNOPs). Simulations with various parameters confirm the efficiency and real-time performance of the developed DZNNM for TVNOPs, indicating its suitability and superiority for solving the shortest path planning problem in real time.

A new hybrid genetic algorithm with tabu search for solving the temporal coverage problem using rotating directional sensors

AbstractOne of the most important problems in directional sensor networks is coverage problem. The coverage can be measured in two ways: positional or temporal. In temporal coverage, the directional sensors rotate periodically round themselves in a repetitive process. Thus, in each time slot, those targets that are positioned within the sensor nodes radius receive their desired coverage. In this model, if a target is left uncovered, it is said that the target has remained in darkness. The main task defined for the temporal coverage model is the minimization of the total dark time for all the targets in the network. This problem has been solved by greedy‐based algorithms in last studies. Greedy‐based algorithms are able to solve the temporal coverage problem in real time. Remember that the performance of greedy algorithms is extremely dependent on the closeness of optimal solution and initial candidates. For this reason, greedy algorithms may obtain local minima due to heuristic search. As far as we know meta‐heuristic algorithms have not been used in past researches to solve such problems. For solving this problem, in this paper two algorithms were developed, GA‐based and hybridized model comprising genetic algorithms and tabu search. A new model was suggested for the chromosome in genetic algorithm. To evaluate the performance of the developed algorithms, they were compared with randomized scenario and greedy‐based algorithm presented in last studies. For better comparison, several parameters, including total dark time, number of sensors, number of targets, sector angle, sensing range were taken into account. The results obtained from the comparison of the algorithms indicated that the developed algorithms are effective in solving the temporal coverage problem in terms of minimizing the total dark time of the targets.

Convex Approach to Real-Time Multiphase Trajectory Optimization for Urban Air Mobility

Electric vertical takeoff and landing technology has emerged as a promising solution to alleviate ground transportation congestion. However, the limited capacity of onboard batteries enforces hard time constraints for vehicle operations in a complex urban environment. Therefore, a constrained, real-time trajectory optimization method is necessary to enable safe and energy-efficient vehicle flight. Unfortunately, most existing trajectory optimization methods suffer from low computational efficiency, unpredictable convergence processes, and strong dependence on a good initial trajectory. In this paper, we employ a successive convex programming approach to solve the multiphase minimum-energy-cost electric vertical takeoff and landing vehicle trajectory optimization problem for urban air mobility applications. The main contribution is the development and validation of two novel convexification methods that find approximate optimal solutions to the multiphase nonlinear trajectory optimization problem in real time by solving a sequence of convex subproblems. Specifically, the first method transforms the original nonlinear problem into a sequence of second-order cone programming problems through a convenient change of variables and lossless convexification, while the second approach achieves a similar goal without the need for control convexification. The resulting convex subproblems can be solved reliably in real time by advanced interior point methods. The proposed methods are demonstrated through numerical simulations of two different urban air mobility scenarios and compared with the solution obtained from GPOPS-II, a state-of-the-art general-purpose optimal control solver. Our proposed sequential convex programming methods can obtain near-optimal solutions with faster computational speed than GPOPS-II.

Penerapan Metode Vikor dalam Pemilihan Bibit Unggul Pohon Karet

Rubber plant is one of the plantation commodities that has an important role in economic activities in Indonesia. The need for rubber seeds continues to increase in line with the increase in the area of ​​smallholder rubber plantations and the government. Quality seeds are a community need in developing rubber plantations in Indonesia from year to year. The first step to get good rubber seeds is that rubber farmers need to use quality rubber seed planting material and are able to produce high latex. Given the very importance of seeds in determining quality rubber repair. With a Decision Support System with the vikor method to build a system that has the ability to be able to assist farmers in choosing superior rubber tree seeds with a system that is able to provide problem solving skills and communication skills for problems with semi-structured and unstructured conditions. This study uses the vikor method because this method is suitable for use in most real-time problems such as making decisions to find quality rubber tree seeds so that this research can be useful for rubber tree farmers in improving the quality of superior seeds.

Education Model for Paediatric Palliative Care

Introduction: Meeting the palliative care needs of children and their families is complex and challenging. For countries such as Australia, whose relatively small paediatric palliative care population is dispersed across a very large geographical area, one challenge is maintaining a skilled workforce in regional, rural, and remote areas, where, when compared to major cities, there are fewer resources, and the workforce is often transient. Methods: The Quality of Care Collaborative Australia (QuoCCA) Pop-up Model of Education was used to provide palliative care education to health professionals across three geographical locations and facilities, to facilitate a 2,500-km transfer of a child with complex palliative care needs from a tertiary hospital to the remote family home on Country. Results: Each Pop-Up provided effective education to facilitate the successful transfer of the child to the next hospital location. Over 18 months, three Pop-Ups occurred. Relational learning and real-time problem solving enabled health professionals to build confidence and capacity to successfully transfer the child from the regional hospital to the remote family home. Implications: The QuoCCA Education Pop-up Model is a feasible method to deliver timely access to speciality education. The model can be successfully applied in multiple settings. Keywords: paediatric, palliative care, integrated health care delivery, learning, rural, education models.

Research on Improved YOLOv5 Vehicle Target Detection Algorithm in Aerial Images

Aerial photoelectric imaging payloads have become an important means of reconnaissance and surveillance in recent years. However, aerial images are easily affected by external conditions and have unclear edges, which greatly reduces the accuracy of imaging target recognition. This paper proposes the M-YOLOv5 model, which uses a shallow feature layer. The RFBs module is introduced to improve the receptive field and detection effect of small targets. In the neck network part, the BiFPN structure is used to reuse the underlying features to integrate more features, and a CBAM attention mechanism is added to improve detection accuracy. The experimental results show that the detection effect of this method on the DroneVehicle dataset is better than that of the original network, with the precision rate increased by 2.8%, the recall rate increased by 16%, and the average precision increased by 2.3%. Considering the real-time problem of target detection, based on the improved model, the Clight-YOLOv5 model is proposed, by lightweighting the network structure and using the depth-separable convolution optimization module. After lightweighting, the number of model parameters is decreased by 71.3%, which provides a new idea for lightweight target detection and proves the model’s effectiveness in aviation scenarios.

Kriging-based Model Predictive Control for Lower-limb Rehabilitation Robots

Model predictive control (MPC) has emerged as a predominant method in the realm of control systems; yet, it faces distinct challenges. First, MPC often hinges on the availability of a precise and accurate system model, where even minor deviations can drastically affect the control performance. Second, it entails a high computational load due to the need to solve complex optimization problems in real time. This study introduces an innovative method that harnesses the probabilistic nature of Gaussian processes (GPs), offering a solution that is robust, adaptive, and computationally efficient for optimal control. Our methodology commences with the collection of data to learn optimal control policies. We then proceed with offline training of GPs on these data, which enables these processes to accurately grasp system dynamics, establish input–output relationships, and, crucially, identify uncertainties, thereby informing the MPC framework. Utilizing the mean and uncertainty estimates derived from GPs, we have crafted a controller that is capable of adapting to system deviations and maintaining consistent performance, even in the face of unforeseen disturbances or model inaccuracies. The convergence of the closed-loop system is assured through the application of the Lyapunov stability theorem. In our numerical experiments, the exemplary performance of our approach is demonstrated, notably in its capacity to adeptly handle the complexities of dynamic systems, even with limited training data, underlining a significant leap forward in MPC strategies.

Inappropriate YouTube content detection and classification by using proposed novel auto-determined k-means clustering and PDBRNN architecture

To ensure a safe and pleasant user experience while watching content on YouTube, it is necessary to identify and classify inappropriate content, especially content that is inappropriate for children. In this work, we have concentrated on establishing an efficient system for detecting inappropriate content on YouTube. Most of the work focuses on manual pre-processing; however, it takes too much time, requires manpower support, and is not ideal for solving real-time problems. To address this challenge, we have proposed an automatic preprocessing scheme for selecting appropriate frames and removing unwanted frames such as noise and duplicate frames. For this purpose, we have utilized the proposed novel auto-determined k-means (PADK-means) algorithm. Our PADK-means algorithm automatically determines the optimal cluster count instead of manual specifications. By doing this, we have solved the manual cluster count specification problem in the traditional k-means clustering algorithm. On the other hand, to improve the system’s performance, we utilized the Proposed Feature Extraction (PFE) method, which includes two pre-trained models DenseNet121 and Inception V3 are utilized to extract local and global features from the frame. Finally, we employ a proposed double-branch recurrent network (PDBRNN) architecture, which includes bi-LSTM and GRU, to classify the video as appropriate or inappropriate. Our proposed automatic preprocessing mechanism, proposed feature extraction method, and proposed double-branch RNN classifier yielded an impressive accuracy of 97.9%.

The influence of feature grouping algorithm in outlier detection with categorical data

Outlier mining has become a rapidly developing domain over the recent years with increasing importance in the fields like banking, sensor networks, and health care. In general, anomaly detection methods are compatible with numerical data and ignore categorical data. However, in real-time problems, both numerical and categorical data are to be considered to obtain accurate results. There are several methods available for the outlier detection of high dimensional data in numerical data. In this paper, a feature grouping algorithm for anomaly detection is proposed that considers the categorical data also. This algorithm correlates the features of categorical data and forms feature clusters and detects the outliers. The features are assigned feature weights based on their levels of appearance and the outlier scores are determined. The performance of the feature grouping algorithm is then compared with the traditional algorithms like LOF and Isolation Forest algorithm and state-of-the-art methods like WATCH on UCI datasets. From the experimental evaluation of the results obtained, it is found that the proposed algorithm is comparatively better than the existing algorithms for categorical data.

Hydro Sight: Groundwater Prospector

Access to groundwater is crucial for the sustenance of communities, particularly in key regions of West Bengal such as Midnapore, Birbhum, Purulia, and other areas. A significant portion of the population relies heavily on groundwater for various essential needs, including drinking water and daily chores. Hydro Sight is a project aimed at revolutionizing groundwater prospecting using data analysis techniques. Our initiative leverages data provided by the State Water Investigation Directorate (SWID), Government of West Bengal. Using machine learning techniques, we will generate compelling visual representations of data. Hydro Sight isn't just about technology; it's about working together to solve a real-time problem.

Path and cost optimization using genetic algorithm: an application perspective

Genetic Algorithm is an optimization technique inspired by nature. The technique has been used by scientists and engineers for real-life search and optimization problems. This work makes use of the genetic algorithm for the solution of the traveling salesman problem. This work focuses on real-time problems; the algorithm is used to find the optimum path for sales travelers inside the Khyber Pakhtunkhwa Province of Pakistan. The solution provides the shortest distance between the cities to be traveled and gives the optimal route. The coding is done in Python-3 and software is developed for the traveling salesmen, where the salesmen can select the cities, they want to travel, and the software will provide the optimal path and the distance.

A quantum moving target segmentation algorithm for grayscale video based on background difference method

The classical moving target segmentation (MTS) algorithm in a video can segment the moving targets out by calculating frame by frame, but the algorithm encounters a real-time problem as the data increases. Recently, the benefits of quantum computing in video processing have been demonstrated, but it is still scarce for MTS. In this paper, a quantum moving target segmentation algorithm for grayscale video based on background difference method is proposed, which can simultaneously model the background of all frames and perform background difference to segment the moving targets. In addition, a feasible quantum subtractor is designed to perform the background difference operation. Then, several quantum units, including quantum cyclic shift transformation, quantum background modeling, quantum background difference, and quantum binarization, are designed in detail to establish the complete quantum circuit. For a video containing 2m\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$2^{m}$\\end{document} frames (every frame is a 2n×2n\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$2^{n} \ imes 2^{n}$\\end{document} image with q grayscale levels), the complexity of our algorithm is O(n+q)\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$(n+q)$\\end{document}. This is an exponential speedup over the classical algorithm and also outperforms the existing quantum algorithms. Finally, the experiment on IBM Q demonstrates the feasibility of our algorithm in this noisy intermediate-scale quantum (NISQ) era.