Yoshiro Hada

A robust room-level localization method based on transition probability for indoor environments

Several scene analysis methods [1, 2] such as location fingerprinting have been proposed to execute room-level localization with high accuracy. However, these methods estimate locations wrongly when received signal strength (RSS) observed at a target users current position is similar to RSS in another place. By utilizing the heuristic of passing through a passable boundary point (for example, an entrance door) when moving between rooms, we introduce a technique for evaluating the distance between the current position of the target user and the boundary point by comparing RSS around the boundary point with RSS at the user in order to reduce the number of erroneous judgments. Experiments conducted in office environments confirmed that our proposed technique achieved an accuracy rate of 97.1% in localizing the target user.

Hybrid positioning system combining spatially continuous and discrete information for indoor location-based service

Positioning systems have gained importance in situations where various location-based services have been installed. Although the combination of Wi-Fi positioning with inertial navigation is considered to be accurate for indoor use, it is important to avoid increasing the error when using estimated positions for location-based services. In this paper, we propose a positioning system that combines several positioning methods with different characters to avoid increasing the error. Demonstration experiments that show the effectiveness of the proposed system in an exhibition hall are also described.

Real-Time Deadlock-Free Navigation for Multiple Mobile Robots

This paper proposes a practicable navigation method for multiple mobile robots in a realistic environment. In the past, many navigation methods have been developed, however, they were often limited to a single robot or sometimes assumed robots with special mobility such as holonomic ones. From the viewpoint of practical applications it is indispensable that the number of robots is arbitrary, general shape and mobility of them are allowed. In this study, a deadlock-free navigation for non-holonomic mobile robots in a practical environment is given. In the proposed scheme, states of the environment and robots are feedback in real-time and global path planning is cyclically executed. It enables an adaptation to a changing environment and a reliable deadlock-free navigation of multiple robots. The real-time online path planning is performed by an efficient A* search in Configuration Spaces representing the robots and environment. The proposed method is tested by several simulations which represent typical complicated navigation situations. As a result, the effectiveness of the methods is verified.

Recognition of hand action using body-conducted sounds

Several methods of recognizing hand actions by examining the vibrations conducted through the body from muscular activity (we call them “body-conducted sounds” in this paper) were proposed in previous works. However, they did not consider the transfer characteristic of body-conducted sounds. In this paper, we propose a method for hand action recognition that extracts the main frequency elements of body-conducted sounds using the Mel-Frequency Cepstrum Coefficient (MFCC) and divides the hidden states of Hidden Markov Models (HMMs) based on the MFCC. The results of experiments show that our method makes it possible to correctly recognize 95% of hand actions on average.

Development and field test of novel two-wheeled UAV for bridge inspections

This paper presents the development and field test of a novel unmanned aerial vehicle (UAV) for bridge inspections. The proposed UAV, which consists of a quadrotor, a cylindrical cage installed in it, and two spokeless wheels freely rotating around the cage, can climb and run on the bridge surface. Its structure improves durability and reduces air resistance compared with conventional UAVs with a cage or wheels. These advantages expand the uses of the UAV in various fields. This paper evaluates the effectiveness of the proposed UAV in real-world bridge inspection scenarios. We tested basic locomotion and measured air resistances. Experimental results on bridges demonstrated the ability to inspect various locations on bridges that are difficult for human inspectors to access.

Power assist method for a nonholonomic mobile robot using both ZMP criterion and impedance control

In this paper, we describe a power assist method for a nonholonomic mobile robot designed to carry out office-oriented services. We developed a new mechanism for obtaining an approximate measurement of the zero moment point (ZMP) using three single-axis force sensors so that a robot pushed from any direction by a human can detect the direction and magnitude of the external force. We also propose impedance control to generate translational and rotational motion in accordance with ZMP displacement. This control method achieves a function whereby a robot that has been given an external force from a direction differing from its translation direction will follow the direction of the external force while rotating. The robot can also return automatically to its initial pose once the external force has been released. We show though experiments that our power assist method is effective in improving the operability and usability of an office-oriented robot.