More robots are involved with humans, such as cooperative robots. When providing services based on user attributes (name, authority, location information, etc.) or authentication (face, fingerprints, etc.), developers must be careful to protect personal information. Therefore, we developed and released a package called “sros2_oidc”. The app adopts the OIDC (OpenID Connect) as an authentication protocol. It is one of the standard authentication protocols for online services. And combined with SROS2, personal information is sent through a secure route to the end of the robots.

Wiping the surface of an object with a robot is necessary to automate various tasks such as cleaning and polishing. However, it is very difficult to move contact points while maintaining the appropriate force with position control. It is because the force response when contact with the object varies greatly with slight differences in the positional relationship, it is greatly affected by errors in the shape measurement and positional errors in the robot's motion. To achieve this, force control that directly handles the force response is necessary. In this study, we used bilateral control-based imitation learning to learn wiping motions for several types of bowls as objects with various 3D curved surface shapes, and verified their performance.

We are conducting research and development of Hakoniwa, a virtual simulation environment in the age of IoT and cloud robotics. This research aims to deploy the core functionality of Hakoniwa for simulation of ROS 2 applications. The proposed method mainly consists of Docker and Unity, and allows engineers to easily try out robot development as many times as they wish on multiple platforms independent of OS environment.

To improve the quality of life (QOL) of the elderly, we implemented a counseling robot that responds according to the results of AI-based QOL estimation during the interaction. Based on the results of a one-week interaction experiment with young and elderly participants, we concluded that the proposed method has the potential to improve the mental aspect of QOL, and the need for validation through large-scale experiments became clear.

For virtual usability evaluation of universal design, our previously-developed grasp synthesis method for human hand model was extended by employing soft finger model. Our previous method treated a hand model as a hard finger that assumes point contact with friction, which was not able to synthesize a grasp with two-point contact that often observed in the real world. Therefore, we incorporated the judgment of graspability based on the soft finger model and confirmed the synthesis of multiple types of plausible grasps including two-point contact through the simulation.

This paper describes the elemental technology of a system that discriminates whether or not sputum is accumulated (whether or not suction is necessary) based on auscultation sounds using edge devices such as microcomputers, in order to support endotracheal suctioning (a procedure to suction sputum accumulated in the trachea), which is one of nursing tasks. Since a typical FFT library for general microcomputers is used, a long-span FFT is required to process a single lung sound. On the high-performance microcomputer SPRESENSE, auscultated lung sounds were FFTed over a long span (approximately 4.16 seconds), and the frequency spectrum was used to discriminate normal from abnormal using a neural network model that was machine-learned on the computer beforehand. As a result, the discrimination accuracy was confirmed to be relatively good, about 96%.