Some electric-powered wheelchairs are recently redefined as personal mobility devices. Their users are not only elderly or handicapped people, but also passengers with large baggage or pedestrians going from station to destination, i.e., last-mile transport. Consequently, people with different operation skills and expectations on personal mobility would become new users of this kind of devices. Safe and comfort travel in human co-existing environment such as sidewalks and airports is a social expectation for personal mobility. In order to realize this, understanding the operation skill of each user by a practical and simple method is essential. This paper thus introduced a skill level classification method by machine learning using only joystick data as input. In order to determine the number of skill level clusters, basic 26 features of joystick operation data are used for unsupervised clustering (single-linkage). We then made evaluation indexes by using speed, speed control, and direction control. For a five-level classification by using gradient boosting as supervised learning, we achieved a 67% accuracy (tolerance: 0) and a 98% accuracy (tolerance: 1). Further analysis of the feature importance of gradient boosting revealed key features to a good operation. Results also show that skill level differed among people with different driving experiences.

We constructed a proof of concept demo system for IoT services with remote controllable IoT devices to show appropriate network architecture to assist the service. The demo exhibits how our proposed approach for supporting an application service for touch feeling can improve the transmission of actuation information emitted from a haptic device in high frequency. Our approach uses multiple paths and compression technique by grasping the characteristics of the tactile information obtained from haptic devices. We constructed our demo by partially implementing data-plane of the proposed system and exhibited the effectiveness of our approach.

This paper focuses on a remotely controlled feedback-based robotics system in a wide area network. Our use case is that a remote cloud server will aggregate advanced information processing functions for data measured by sensors and feedback to robotics devices for actuation. Network performance and Quality of Service (QoS) affects the stability of control. Therefore, we propose a proxy-based network function that allows the coexistence of service traffic flows for a robotics feedback control system with other services. We compared our approach with a case that did not apply our method by using simulation and showed the effectiveness of our proposed scheme.

Quality of experience that is perceived by customers directly affects reputation web service providers. Major web service providers locate intermediate computing resources between their contents servers and customers that assist quality enhancements of their service for their customers. For services using Internet of Things (IoT), there are strong interests in controlling intelligent IoT devices, especially robotics devices, through network. However, network status affects the quality of this remote control service. We propose network assisting functions and supporting architecture for such services. In our proposal, we spin off parts of IoT devices or server functions from their chassis and locate them on network located computing resources as software-based functions. We demonstrate the effectiveness of our proposal using a remote controlled robotics service.

We investigate relationships between application level Quality of Service (QoS) for Internet of Things (IoT) and that of underlying networks. We particularly focus on intelligent IoT devices, such as drones and robots, and network technologies supporting them. We assume our target use cases as IoT services controlled by remote servers or terminals operated by human. In this service, delayed information causes unstable behavior of the IoT devices. In this paper, we propose that delayed information that is not suitable for the feedback of the device control shall be discarded in a network. With our prototype systems and a manipulator as an example of the intelligent IoT device, we show effectiveness our proposed scheme and mention the need of Quality of Control (QoC) instead of QoS for these systems.