In Japan, it is serious problems such as increasing nursing care of handicapped people requiring and the aging of caregivers which is caused in a super-aging society. One of the main causes of being classified as persons requiring nursing care include joint diseases such as rheumatoid arthritis and bone fractures due to osteoporosis. Early detection and treatment of these diseases are considered important. Rheumatoid arthritis and osteoporosis are generally diagnosed by simple X-ray examination. However, there are problems with radiographic diagnosis by physicians, such as lack of objectivity and reproducibility of diagnosis, and increased workload on the radiologists. To solve these problems, a Computer-Aided Diagnosis (CAD) system is being developed. Because the CAD system may use the results of quantitative computer analysis, it is expected to improve the reproducibility and accuracy of diagnosis and reduce the burden on physicians. Therefore, this paper proposes a segmentation method of phalanx region from CR images to develop a CAD system. The proposed method is based on HRNet + JPU + U-Net. The proposed method was applied to 101 cases of X-ray images, and mIoU=0.897 was obtained. Experiments for segmentation from images confirmed the usefulness of the proposed method by improving the extraction accuracy of the boundary of the phalanx region.
In this paper, a modified super-twisting active disturbance rejection control (MSTADRC) and a nonlinear observer (NOB) are used to implement a sensorless speed control for surface mounted permanent magnet synchronous motor (SPMSM), which has a complex internal structure and is characterized by nonlinear and variable parameters. A new reaching law is designed in super-twisting sliding mode control (STSMC). We proposed the MSTADRC via STSMC and active disturbance rejection control (ADRC). NOB has been selected for estimation of position or angle value. Finally, the validity and effectiveness of this approach are illustrated by simulation.
An automated material handling system (AMHS) is part of a production system that transports products from one machine to another for manufacturing processes. While conveyor belts have been commonly employed as AMHS, the increasing demand for enhanced performance has led to the emergence of advanced AMHS based on linear motors. A feature in these advanced AMHS is that they often allow non-uniform transfer speed setup, i.e., transfer speeds are allowed to be different by individual sections within the production system. This characteristic introduces an opportunity in determining the optimal transfer speed vector for the maximum productivity. In this paper, we propose an algorithm for optimizing the transfer speed vector using an analytical throughput evaluation method. The effectiveness of the algorithm is demonstrated through an example involving five machines.
Artificial Intelligence (AI) gives a new way to engineering practices. The inverse kinematics (IK) solutions of a robot manipulator could be done by conventional ways such as geometric, algebraic, or Jacobian methods, which have drawbacks. The different practice that only a few points (training samples) of the end effector are recorded can be planned to operate a manipulator by AI. This practice will no longer need cumbersome IK calculations to plan a manipulator's motion. In other words, IK analysis is approached by the Artificial Neural Networks (ANN). More training samples and hidden neurons are, better the IK function fitted by ANN performs. However, with a high number of samples and hidden neurons, the procedure becomes impractical. This paper attempts to provide a mathematical framework for a reasonable number of ANN training samples for acceptable operations. The study was applied to a 3 degrees-of-freedom (DOF) manipulator. Mathematical bound estimates, knowing trained points can be almost exactly approximated, are derived between trained and untrained points. Through simulation studies, errors by bound estimates and by ANN were compared. Also, the smallest sample set could be drawn; 20 times greater sampling than an allowable tracking error size was found.
In this study, we designed a data-driven fault detection and identification method for time-varying nonlinear systems using the Koopman operator. Koopman operator is an infinite-dimensional linear operator that transforms a nonlinear dynamical system. In this paper, Weighted Window Extended Dynamic Mode Decomposition (WW-EDMD) is used to obtain the Koopman operator through a recursive procedure to reduce the computation time and memory usage. The forgetting factor is implemented to enhance the fault detection ability, weighting the latest data in the WW-EDMD framework. When the column vector norm of the matrix $B$ is less than the designed threshold, the fault is detected and estimated. Through the data collection period, the performance of the obtained Koopman operator model is confirmed by comparing the true state of the model. Numerical simulation results show that the proposed method has better fault detection ability for time-varying nonlinear systems than window extended dynamic mode decomposition.
In order to overcome the shortcomings of the traditional process-oriented aero-engine simulation software, an object-oriented simulation software platform was developed to model the specific components of the variable cycle engine, and the configuration of the variable cycle engine could be built freely by users. Three common variable cycle engines are set up in this paper. Considering their different structures, given the same pressure ratio and bypass ratio and other parameters, the potential of the three variable cycle engines cannot be brought into play. In this paper, intelligent optimization algorithm is adopted to adjust parameter optimization space for cycle parameters and non-design point components given the same design point. The performance of the proposed engine is optimized under the conditions of ground take-off point and high altitude cruising point. The optimization results show that the comprehensive performance of ACE is better than that of FLADE VCE and CDFS VCE.
Cleft lip is one of the most common birth defects. Several operations are performed to form a natural lip. A problem with these operations is that the criteria for the facial symmetry are unclear. Based on this background, we propose a method for evaluating the facial symmetry using 4D point cloud data. We evaluate the facial symmetry in two ways. One is based on the temporal changes in the face landmarks. Corresponding points are searched using two local geometric descriptors. The results are approximated to the 3D lines and these slopes are compared to obtain the left-right difference of the movement. The other is based on the center of gravity. The lip part is extracted from the point cloud and divided into subregions. The centroid coordinates are then calculated for each subregion and compared to obtain the left-right difference of the facial structure. The experiment is performed using artificially generated 4D data. As an experimental result, it is shown that our method can find point correspondences with smaller error than comparative methods.
High-speed rail (HSR) increases undoubtedly the economic productivity as it can promote a sustainable economic development. In order to improve the broadband services and train operational safety, a set of calculations about the feasibility and performance summary on HSR communication link must be provided. This study proposes a designed power link budget to be utilized in 900 Mhz as the frequency spectrum of GSM-R (The Global System for Communication for Railway). The nonstationary and diverse mechanism of HSR make the prediction of path loss as the most prominent factor in HSR channel modeling very difficult to conduct. An optimal link budget must comprise a total calculation of all parameters in which the factor of path loss must be included during the signal propagating between the base station and HSR. In this study, the prediction of path loss values is achieved from the utilization of optimization technique which differ from other existing path loss prediction. With some additional parameters in the link budget, the Received Signal Level (RSL) has shown an acceptable result in the HSR's side. Therefore, HSR network designers can utilize the link budget for the deployment of base stations, frequency set up and coverage area planning.
This paper presents a novel three-DOF robotic finger that enables fast tapping motion applicable in piano playing and keyboard typing, which consists of metacarpophalangeal(MCP), proximal interphalangeal(PIP), and distal interphalangeal(DIP) joints. In previous study, we developed Twisted Round-belt Actuator(TbA) capable of producing contraction force by twisting a small-diameter round-belt having high elasticity, and empirically derived a contraction force model applicable only to two-DOF robotic finger. This study, therefore, derives such force model available for an extended three-DOF finger mechanism that is containing Variable-pitch Screw Module(VpSM) newly designed in this paper. This module is able to eliminate irregular twist phenomenon resulting in unexpected discontinuous movement in finger joints. Thus, we first describe the detail mechanism of the three-DOF robotic finger and indicate great advantage of the finger, which is obtained by combining the TbA and VpSM. Finally, we demonstrate a new force feedback method based on fingertip force estimation without any force/pressure sensors, in which the fingertip force can be accurately estimated by using mathematical model of the contraction force by TbA with VpSM.
In the present work, a new iterative algorithm (CGLS) is proposed to solve the constraint solutions of period-coupled operator equations. The constraint solution here we are dealing with is the Hamiltonian solution. When the equations of the studied period-coupled operators are consistent, it is proved that for given any initial complex matrices, the constrained solutions can converge to exact solutions. Otherwise, when the studied period-coupled operators are inconsistent, the minimum norm constraint solutions can also be calculated by selecting any initial matrices. Finally, some numerical examples are given to illustrate the effectiveness and superiority of the new methods.
This paper proposes a new methodology for deriving a point-based dimensionally homogeneous Jacobian, intended for performance evaluation and optimization of parallel manipulators with mixed degrees of freedom. Optimal manipulator often rely on performance indices obtained from the Jacobian matrix. However, when manipulators exhibit mixed translational and rotational freedoms, the conventional Jacobian's inconsistency of units lead to unbalanced optimal result. Addressing this issue, a point-based dimensionally homogeneous Jacobian has appeared as a prominent solution. However, existing point-based approaches for formulating dimensionally homogeneous Jacobian are applicable to a limited variety of parallel manipulators. Moreover, they are complicated and less intuitive. This paper introduces an extended selection matrix that combines component velocities from different points to describe the entire motion of moving plate. This proposed approach enables us to formulate an intuitive point-based, dimensionally homogeneous Jacobian, which can be applied to a wide variety of constrained parallel manipulators. To prove the validity of proposed method, a numerical example is provided utilizing a four-degree-of-freedom parallel manipulator.
In a power system, low-quality electricity can become a serious cause of equipment degradation, power outages, or lethal accidents. Quality indices that have received attention from many researchers are voltage and frequency deviations. Since renewable energy can be considered a cause of power instability in the system, its widespread adoption may have a significant negative impact on power quality. To address this issue, particle swarm optimization and a prediction model based on load profiles are utilized to mitigate voltage and frequency deviations in a multi-area power system. Verification is conducted using real-world data obtained from a university campus in Thailand, including power generation from solar photovoltaic cells and daily power consumption. Other prerequisite assumptions include the installation of battery energy storage systems and flexible AC transmission system devices that enable power-sharing between interconnected feeders. By adjusting the tie-line synchronizing coefficients between areas, the result demonstrates that the proposed strategy effectively improves the power quality in the multi-area power system.
When lung cancer is suspected, a CT scan of the chest is widely used as a method of precise examination. However, the number of CT images obtained in a single examination is enormous, placing a heavy burden on the physician who reads the images. In addition, there is concern about the possibility of undetected lesions due to differences in the skills and experience of the reader. Therefore, computer-aided diagnosis systems have been introduced to reduce their workload and undetected lesions. When physicians make a diagnosis, they consider not only CT images but also information about the patient. Therefore, attempts are being made to improve the accuracy of diagnosis by mimicking this process using artificial intelligence. In this paper, we propose a model for identifying nodular shadows using deep learning, aiming to improve the accuracy of diagnosis by introducing medical record information in addition to image information. Normal tissue is divided into three classes: branched vessels, thin vessels, and round vessels, and a total of four classes are classified, including abnormal tissue. Experimental results show that the accuracy of nodular shade discrimination is improved when medical record information is added.
“Read to a Dog” is a program operated in the United Kingdom, the United States, and New Zealand to induce children's interest in reading through children's reading to dogs. This program is a type of animal-assisted therapy (AAT) that uses pets to heal psychological problems, and it can reduce reading rejection and induce interest in reading by providing an environment where children can read confidently and feel comfortable. This study aims to develop a dog-type robot that can be operated in the library based on the research contents of the corresponding “Read to a Dog”. The purpose of this study is to stimulate children's interest in reading through a reading activity support robot, while giving them with an emotional experience of artificial intelligence through robots. Through robot use environment analysis and user analysis, the appearance shape of the robot is designed and the service requirements are defined. In addition, for touch interaction for emotional communication between children and the dog-type robot, this paper proposes a touch sensor technology using a pressure sensor and Micro Electro Mechanical Systems (MEMS) microphone, which can reduce hardware complexity and detect a wide range of touch of a robot.
This study proposed a new approach to learning control of a mobile robot system with three wheel based voice recognition for smart factory. The motion control technology was applied by back propagation based on multi-layer neural network. A experiment result has been given in which some artificial assumptions about the linear and the angular velocities of mobile robots from recent literature are dropped. In this study, we proposed a new approach to real time control of the position and velocity for mobile robot system. The simulation results were proposed, which confirmed the effectiveness of the proposed control algorithm. Moreover, practical experimental results of the real time control were reported with several real line constraints for mobile robots with three wheels.
This paper proposes an active noise control (ANC) algorithm that enable to use in a sparse system environment, An adaptive filter, which is one of the methods for estimating the inverse channel, was used to perform active noise control. The filtered-x algorithm was employed through the adaptive filter to address the issue of secondary path distortion in active noise control. For estimating the inverse channels in the sparse system, the $l_{0}$ -norm was employed, and a newly analyzed mean square deviation (MSD) was applied to calculate the optimal step size that enhance convergence performance. The modified reset algorithm was also applied to ensure that the adaptive filter can track the inverse channel properly even in situations where the system suddenly changes, Various simulations show that the proposed ANC algorithm has better convergence performance than the other algorithms that used in sparse system, In addition, the tracking performance in system sudden change situation shows that the proposed algorithm estimates the inverse channel properly.
This paper proposes a novel model architecture, the Hamiltonian neural network (HNN)-Transformer, which capitalizes on the strengths of both Hamiltonian neural networks and Transformers to effectively model and predict the behavior of physical systems. The proposed structure is designed to solve the problem that arises over time when predicting the state of a 2D robotic system. The HNN component of our model makes it possible to incorporate known physical laws into the learning process, while the Transformer component enables effective processing of sequentially input data. The performance of the proposed method was confirmed through simulation, and it was confirmed that this novel model is possible to leverage the strengths of both HNNs and Transformers to achieve improved performance than the existing independent HNN and Transformer on the challenging task, respectively. This suggests that the integration of physical understanding and sequence processing is a promising direction for modeling complex dynamic systems.
Velocity and position control for metro trains is typically achieved by classical control methods (PID, etc). Challenges in this control problem include imprecise position sensing, time delay, and external disturbances due to weight changes, curves, and slopes of the rails. In order to achieve acceptable stop position of the trains at each station, the controller design often involves individual gain tuning for each sections in the route, which consumes much time and effort. As a means to reduce the effort, reinforcement learning approach is looked into for train control. Automatic Train Operation (ATO) simulator capable of realistic simulation of train dynamics along the Line 5 in Seoul Metro is used to investigate the feasibility of this approach. Results are discussed from the perspective of practicality.
Data-enabled Predictive Control (DeePC) allows controlling dynamic systems soley based on its input/output data. This approach is based on behavioral theory, which guarantees precise prediction of the output for given input as long as the collected input data satisfy Persistency of Excitation (PE) condition and the system is linear time invariant. In practice, however, DeePC faces to control nonlinear dynamics and it is necessary to investigate whether there is a preferred way of collecting input and output data for DeePC besides the PE condition. This paper investigate the issue using an Automatic Train Operation (ATO) simulator that represents existing metro train control systems including time delays and nonlinearities. We implement DeePC using two different datasets to control metro train. Comparison and discussion are provided.
