Entropy sources (e.g., physical phenomena) are essential for true random number (TRN) generation. An unpredictable random number (URN) generator was previously proposed, which uses processor internal registers as its entropy sources. Another study proposed to integrate a linear feedback shift register (LFSR) in a processor, and sample it to generate a URN sequence. The entropy source of this URN is the fluctuation of sampling period. The current study proposes to use weather data as the entropy source for URN generation, where the sampling period is modified by the wind direction data. The derived URN sequences passed the Diehard test when the least sampling period β > 29 with a 32-bit LFSR. It also passed the NIST test when the weather data were accessed with an appropriate hash function when β was 32.

A previous study presented a random number generator based on the fluctuation in the sampling interval of an internal linear feedback shift register (LFSR). They reported that the derived random numbers passed the Diehard test; however, the design guidelines were not specified. The previous study examined two configurations of LFSR, which are far from satisfactory to determine the appropriate length and feedback polynomial. Moreover, the minimal sampling period was not known. The present study focuses on these aspects and elucidates the requirements of LFSRs to generate high-quality random numbers. Extensive experiments were conducted, and the following results were derived. (1) The selection of the characteristic polynomial of an LFSR affects the quality of random numbers. (2) The length of an LFSR should be 48 bits or greater with a constant sampling period. (3) The sampling period should be 32 cycles or larger. (4) The quality of random number sequence is considerably improved by a fluctuating sampling period. (5) The properly designed random number generator passes the NIST test.

It is important to protect the intellectual property (IP) of embedded software, which contains precious know-hows and trade secrets. As hardware is more difficult to analyze than software, it is possible to protect IP by implementing it in hardware. It is particularly easy to port software to hardware when both the software and hardware are written in the same language. Some recent soft processors have been written in a high-level programming language (e.g., C), which can be synthesized by high-level synthesis (HLS) tools. Sakamoto et al. (2018) implemented a function of a software as a special instruction of a MIPS-based soft processor. They also pointed out that passing arguments and values involves various problems. Iwamoto et al. (2019) and Masanobu et al. (2020) enhanced Sakamoto's method; however, their evaluation results were imperfect and problematic. This study resolves many of these problems and reports more evaluation results than previous works. Detailed evaluation results suggest that the hardware implementation of a function is possible and that the overhead in latency and hardware cost is modest.

Development of a system that can automatically detect appearance defects of piston-ring components of the engine cylinder caused during the coating process. In the proposed system, piston-rings are sent from the feeder to conveyor belt, and an image captured by camera. Subsequently, the image is cut along with the shape of ring into small images. A convolution neural network (CNN) model to classify which piston-ring is a normal and anomaly. Finally, a robot arm is utilized to remove the anomaly piston-ring from the conveyor belt.

In our previous experiment, when a GPU-based computer was used to process images, the system could achieve approximately 90-100% accuracy based on the type of defects. To reduce the costs of system, we study single-board computer (SBC) with Google Edge TPU USB Accelerator to classify images, which exhibits good potential to replace GPU-based processing. Furthermore, this paper also proposes some approaches to improve processing speed when using proposes low-cost SBC platform.

Homography is one of the most fundamental image geometric transformation techniques. Real-time homography can be used in online learning as a support function. In this study, The homography computations were conducted using only one central processing unit (CPU) and a multiple-core CPU with a graphics processing unit and a field-programmable gate array, with or without a conversion table. We also compared the shooting angles to check whether the characters in the image were recognizable after homography. The angle of photography was changed from 10° to 40° in 2° increments, and optical character recognition technology was used for character recognition.

The authors held a training course to learn robotics and AI programming with a drone for junior high school students. This course is a part of the children's science and technology human resources development project in Okinawa Prefecture. Initially, this training course was conducted based on the “robots and control” subject of National Institute of Technology, Okinawa College. A lecture was conducted through team training as a practical engineering design education class. The contents of lecture were software training and demonstration of robot system realization. However, implementing a drone autonomous movement program for junior high school students was challenging. For this reason, we devised a method that uses human movements instead of full-autonomous programs (we used color-object following and human following to control the drone, which will be described later). As a result, we established a training course for making a drone control program for junior high school students in 3 to 4 days. This paper describes and evaluates the system used in the abovementioned training course. The evaluation was based on the ARCS model. We present this teaching system as a practical example of engineering design training in primary education.

In recent years, the percentage of deaths and missing persons due to diving accidents has been increasing in Okinawa Prefecture, with drowning accounting for 70% of the deaths. Therefore, we developed a motion analysis system that to understand the differences in posture between students and instructors to acquire to determine their diving skills. We verified the posture improvement in the group using this system and the group receiving conventional diving instruction, as well as the posture improvement before and after the use of this system. We found that the system had a significant effect on posture improvement. It also promoted an autonomous learning attitude toward skill improvement.

In this paper, we propose predictive advanced intelligent security camera system that can predict a snatching incident before its occurrence on streets. In our past studies, although we have been engaged automated detection of snatching incidents using machine learning techniques and we have been proposed many systems, they are not enough to prevent snatching incidents. Therefore, we are now engaging new studies on prediction of snatching using motion prediction techniques. We are developing PASCAS that can predict the occurrence of snatching from sequential videos observed in criminal scenes. It can predict the situations into criminal or non-criminal scenes. The experimental results show that PASCAS can effectively predict snatching incidents with an accuracy of 94.25%。

One of the impacts of COVID-19 is the exponential growth of online classroom teaching. However, grasping the reaction and understanding of the students in an online classroom is difficult for the teachers. This is especially a problem in Project Based Learning (PBL) courses with practical exercises. This study provides a “place” to share information about problems encountered by each student. Other students can then use this information to perform practical exercises remotely. As a result, we were able to identify bottlenecks in the exercise tasks by creating “empathize with problems” among students.

This paper discusses operation characteristics of discontinuous current mode (DCM) for an ac-dc isolated converter for on-board battery chargers of electric vehicles. The converter discussed in this paper consists of a front-end diode-bridge rectifier, an active energy buffer, and a dual-active-bridge dc-dc converter. The energy buffer circuit absorbs the power pulsation form the single-phase ac line into a small-rated capacitor, and thus, makes it possible to reduce a required capacitance of the dc-link capacitor. This paper explains details of the DCM control method including a technique of the ripple cancel of the inductor current, to realize the power factor correction of the line current and the voltage control of the buffer capacitor at the same time without complex calculations. Then, a reference value of the inductor current and an operation range of the dc-output voltage are discussed. Experimental results obtained by a 220-V, 6.6-kW experimental setup demonstrate a sinusoidal-line current and a controlled buffer capacitor voltage as well as a smoothed dc-output current.

The circuit model is a simple model that can be used for motor control and other applications. Here, we derive circuit model parameters that can take into account the rotor skew, bar temperature, and current distribution in the bar of an induction machine. The proposed method is characterized by the fact that the parameters can be determined only by 2D magnetic field analyses, even for axial structural changes such as skew, and can flexibly respond to design changes. The high accuracy of the model was confirmed from analysis results using the circuit model with the proposed parameters and the measurement data of a test machine.

An interior permanent magnet synchronous motor (IPMSM) is characterized by its high efficiency in a wide range of applications. However, when operating at high speeds, the torque ripple and rotor loss occur because of the spatial harmonics generated by fluctuations in the airgap permeance. This study demonstrates the suppression of slot harmonic components at the magnetic flux density by magnetic wedges made of a magnetic composite material. Results indicated that the torque ripple can be suppressed 8.6% point and the total loss can be suppressed 5.5% by using magnetic wedges. Furthermore, the temperature of the permanent magnet can be reduced because of the rotor loss can be suppressed by magnetic wedges.

Our research focuses on the loss analysis of rotating machines and development of a hybrid-field-type IPM motor.