It is necessary to explore the bottom of the RPV in order to accelerate the decommissioning of the Fukushima Daiichi Nuclear Power Plant. A coupled tendon-driven articulated robot arm, Super Dragon, has been developed to deploy the investigation equipment into the containment vessel. In this paper, we developed an extended Super Dragon with a length of 15.4 m by modifying and integrating the telescopic boom so that it can be mounted on the Super Dragon. Experiments were conducted to raise the telescopic boom vertically and to lift and hold the deployed telescopic boom horizontally. We confirmed the basic movements required to explore the bottom of the RPV.

We are teaching "Fundamental Experiment on Mechatronics". In this experiment, we tried to have students prepare for the experiment by watching a short lecture video, which explains the principle of the experiment, before the day of the experiment together with reading the instruction manual. As a result, we found that while some students were able to gain sufficient understanding, others either did not watch the videos or skipped them, and thus did not fully understand the experiments. In this study, we made a video explaining the principle of the experiment using the “Yukkuri Kaisetsu” slow commentary technique often used on YouTube, etc. The students were asked to watch the video voluntarily, and the results of a questionnaire survey on the ease of understanding were reported.

The pandemic of the COVID-19 has had a major impact on the field of engineering education. Many lectures became online and ensuring the quality of the lectures became a major issue. To improve this situation, we developed a lendable educational tool for mechatronics in on-line lectures. In this paper, its development process, contents and usage experience are reported.

We conduct Project Based Learning training for third-year students in the robotics course of the Department of Mechanical Engineering. In the past two years of training, students have been able to make the robot hand grip an object by manually manipulating the robot arm. However, the problem was that there were no students who could automatically control the robot arm. Therefore, this year, we focused on this issue and conducted practical training by sharing the training between the students in charge of the robot arm and the students in charge of the hand, and they were able to realize a series of movements through mutual communication.

The micro:bit is designed as a microcontroller board for beginners, but it is difficult to use it for visually impaired learners because its programming environment requires a PC. In addition, mobile robots using the micro:bit are equipped with various sensors, but their use is not easy. In this paper, we report on how to improve the accuracy of the sensors in the micro:bit and how to easily use the sensors in the mobile robot micro:Maqueen.

In recent years, the field of robotics and AI has been expanding, and human resource development is necessary. However, there are few teaching materials for beginners. Therefore, we have been developing a series of teaching materials including a textbook, ROS2 and Python development environment using Docker, sample source code and an open robot platform, Happy Mini OP (Open Platform), that is low cost and easy to build for beginners.

In recent years, IoT technology and AI technology have been advancing in society, and the Internet is frequently used to create systems. In particular, IoT technology has become one of the most important issues for training engineers to develop new systems in university education.

In this paper, we report on the development of an environmental monitoring system experiment using IoT technology in the Intelligent Control Engineering experiment of our university. We also report on the results of a questionnaire survey conducted in the class of 2021.

In the future, the size of the domestic robotics market is expected to grow significantly, especially in the service and manufacturing industries, and the opportunities for robotics engineers to play an active role are expected to expand. Our university is training engineers and researchers in the field of mechatronics and robotics who can respond to these needs. In particular, we have introduced competition-style mechatronics education using two-axis drive robots in the experimental subjects implemented in the third year, where students can learn electronic circuit design, control programming, and mechanical design through robot development. In this paper, we describe the contents and issues of the experiments in the third year of introduction.

There are few current robots that are suitable for education. Therefore, the purpose of this study is to develop a small, safe, and secure robot arm for education. The robot arm was made with a 3D printer. Arduino UNO was used as the microcontroller for control. By using MATLAB to control the robot arm, it was possible to control the robot arm using the knowledge gained in the lecture. To evaluate the educational effects of the robot arm developed in this study, we conducted an experiment with 41 students in a lecture on robotics. From the results of the questionnaire, it can be inferred that the students were able to deepen their understanding of the course content and were highly satisfied with the course.

One of the purposes of exploration and inspection of nuclear facilities is to understand the distribution of nuclear materials in reactor. This is because if the elements can be characterized on-site, decommissioning can be carried out efficiently. Therefore, it is necessary to attach a sensor to the tip of the long manipulator to perform on-site elemental analysis. However, the positioning accuracy of the long-reach manipulator is not sufficient due to structural deformation and joint control errors. Therefore, we propose an end-effector that can passively adapt orientation of the measurement device to the subject, and actively generates linear motion to the subject. The end-effector consists of a linear mechanism and a follower mechanism. The end-effector is mounted on a coupled tendon-driven manipulator and operated for an experiment. As a result, we succeeded in performing elemental analysis by LIBS probe even though there was an error in the target posture.

This paper describes a method for discrimination of structures in a nuclear facility by deep learning based on projected images generated from a 3D point cloud using a virtual camera. In order to create a training dataset, we constructed a system that automatically generates both projected and labeled images from 3D point cloud data along a pre-defined trajectory path using a virtual camera. Finally, we reported the training accuracy and discrimination performance by the training dataset created by the proposed method from the point cloud measured at the test plant in the Fukushima Robot Test Field.

In the decommissioning work of Unit 3 of the Fukushima Dai-ichi NPS, it is planned to remove fuel debris. Because of the high radiation levels, fuel debris must be removed using a remote-controlled manipulator. However, at present, there is no manipulator suitable for removing fuel debris. To prepare for designing a manipulator suitable for taking out fuel debris, a simulation of removing fuel debris has done to examine optimum size of the manipulator.

In three-denominational(3D) reconstruction from a group of images, we verify whether scaling the result of 3D reconstruction according to the size of the objects for 3D reconstruction by adding a two-denominational(2D) code whose size is known in advance to the objects. The experimental result showed the accuracy of the scaled reconstructed result was high based on the size of the reconstructed 2D code.

ROS, an open source robot control software platform, is widely used for robot development. Accordingly, robot development engineers are required to have skills in robot control programming using ROS, but in most of the existing ROS-learning contents, it is not possible to confirm that the program and the robot are running in synchronization in the real world at hand. Therefore, we may lose the significance of checking at hand, such as understanding the difference between the virtual world and the physical world. In this study, we propose an e-Learning system for ROS programming that links a lecture-style video with an actual robot at hand. By measuring the latency from the time the video is played to the time the actual robot at hand starts to move, we demonstrated the practicality of the system and summarized the problems and future prospects.

From FY2020, programming education had been started in stages at elementary and junior high schools. Moreover, GIGA School program by MEXT also starts to provide ICT device for all students and network infrastructure for all elementary and junior high school. Authors have been conducted cultivation of volunteers for school education from 2018. Almost 30 senior members were joined the learning program and support for club activities in several schools in Niigata city. Under the COVID-19 disease, it is difficult to conduct these supports in face-to-face. In this paper, practice results of support for club activities by using function of application for on-line collaboration have been reported. The effectiveness of joint use for programming materials and communication between students in different schools have been discussed.

It is expected to innovate the educational process in the field of education, where various cause-and-effect relationships are intricately intertwined. We are certain that co-creative concepts and hands-on learning are essential for education that fosters creative and free thinking. In this study, we propose tensegrity kits as a three-dimensional compositional experience that allows learners to create objects together, thereby promoting creativity through interaction. The tensegrity kit, which we propose, enables us to create structures as a spatial design. In addition, the softness of the object allows learners to experience the deformation of the structure with their hands after the object is completed, which is expected to provide hands-on learning. Furthermore, the tensegrity kit allows wide range of people to enjoy creative activities together and it can be a platform for co-creation. Based on the above, we achieved two types of tensegrity kits: spherical and spine shaped. Thanks to the stronger joints, stabilization of the creative process has been achieved, and tensegrity structures that will not be broken even if squashed with the palm of a hand have been realized, furthermore, it is possible to construct a tensegrity structure with free ends.

This report explains the efforts attempting to develop the engineers through the building the robot as the interface between the real-world and the cyber space in Department of Robotics of Nippon Institute of Technology. Artificial Intelligence (AI) is the most attentional theme in the robotics, and also it is the key technology to develop practical robots moving in the real world with humans in order to recognize the situation surrounding the robots and act properly depending on the conditions. From the point of view of the data flow, the robot is an interface between the real world and the cyber space in the computer mounted on the robots. In order to process the unpredictable, unstructured and complex information in the real world, high-performance computational resource is required. However, there’re insufficient engineers who have the skill to use it. The efforts explained in this report develop the engineers through the development of an intelligent robot from scratch.

Since 2001, Robot-Triathlon, a robot contest for engineering university students in Hokkaido, have been started, and gathered the participants of around 30 teams regularly. Under Covid-19 catastrophe, the contest was canceled in 2020. However, in 2021, it was re-started using an online meeting service. This paper explains its outline, competition rule, operation method of Robot-Triathlon online. Furthermore, the participating teams and robots are introduced and the features of the robots were summarized.

As crops in greenhouses are widely distributed, IoT devices placed near the crops should be stand-alone and modular, and data from the devices are collected over the networks. Smart agriculture requires knowledge of a wide range of fields including electricity, information, and image processing. We have designed an exercise on environmental monitoring and control of a greenhouse toward smart agriculture. The exercise consists of mechatronics, MATLAB/Simulink and system construction exercises. Students learns technology electrical circuit, embedded programing, network communication, and image processing required for smart agriculture. Almost students who took the exercise acquired technology related to micro-computers, image processing technology and control by MATLAB/Simulink, but a few students couldn’t acquire them. In the next school year, authors plan to teach from a little more basic technology.

The goal of this research is to use deep reinforcement learning to acquire dribbling skills for a soccer humanoid robot in an obstacle environment. We also address the problem of learning time, which is one of the challenges in reinforcement learning. Using the constructed simulator, we designed a state transition model that represents the dribbling of the humanoid robot with a Neural Network. By using this state transition model as an environment for reinforcement learning, we succeeded in reducing the learning time and acquiring obstacle-aware dribbling. We confirmed that the learned policy can be transferred to a simulated soccer environment used in RoboCup 2021 Humanoid League. We also tested the trained policy using a real humanoid robot and observed that the robot was able to dribble the ball towards the goal.

The objective of this study is to acquire behaviors for controlling multiple soccer robots using multi-agent reinforcement learning. For the experiments, we constructed a simulation environment for soccer using RoboCup Humanoid League as a reference, and used it as a base for future application to real environments. For the experiments, MA-POCA was used as the reinforcement learning algorithm. We also designed rewards to ensure that the robot learns to play soccer appropriately. As a result of learning, the soccer robot’s policy acquired behaviors necessary for soccer, such as dribbling, passing, and defense. This paper has contributed to the automation of the soccer robot’s behavioral decisions.

We have developed a ROS-based mobile manipulator for restroom cleaning with the capability to recognize garbage of various types (pieces of paper, cups, and liquids) and to select the most appropriate tool from three cleaning tools (suction to hold, vacuuming, and moping) to effectively clean a restroom. Upon deployment at the Future Convenience Store Challenge of the World Robot Summit 2020, we obtained the 1st Place in the Restroom Cleaning Task with an almost perfect score (96%).

For omni wheels used for omnidirectional movement of soccer robots, the shape and number of small wheels affect driving stability and grip. In this research, we will realize the production of small wheels, which was difficult to manufacture by ourselves due to the history of using resin materials, by using a manual tabletop injection molding machine. By eliminating the bearings that have been used so far, we aimed for a more robust wheel. We designed a mold for injection molding and actually manufactured it．We verified the manufacturing method by injection molding with multiple types of resin materials using the manufactured mold．This research makes it possible to propose new small wheel shapes and materials .

In this paper, we described the development of an underwater robot that participated in the AUV division of the underwater competition held at Kobe as a concurrent event of Techno-Ocean 2021. The developed AUV consists of a computer, two kind of power supply, two cameras, and a depth sensor. And, the robot can navigate autonomously by its image processing. In the competition, our team finished in 3rd prize.

Robot competitions are expected to promote innovation of robot development and its utilization, and now in the "with-corona" era, the competition management should consider infection control measures against COVID-19. In this paper, a case study on the Word Robot Summit Disaster Robotics Category is reported, where field layout and additional devices, e.g. virus shield, are examined.

This paper presents the initial system development, integration, and testing of Lab in a Box: a portable, low-cost test and measuring device in aiding Robotics education and development in hybrid or distance learning setups. Powered by a Raspberry Pi, this intends to complement or fully replace bench-top instruments, and provide a scalable platform that suffices for the target Robotics education and development complexity level.

In agriculture, the aging and depopulation of farmers cause the shortages of farmers and manpower. Most of commercialized robots are industrial robots for factory automation, and most of robots for the first industry, agriculture, forestry and fisheries are still under developing. The reasons are cost-efficiency of the robotization, safety of the works using robots, difficulty of outdoor operations, and knowledge transfer problem from farmers to computer, etc. As one of solutions for the problems, robot technology into the agriculture is expected to contribute to the laborsaving, improvement of production, production line automation, and also the management toward smart-agriculture. We organize the Tomato-harvesting-robot competition to offer the research field and welcome researchers into agricultural robotics.