The conventional quadrotor with non-tilting parts has only 4 DoF (Degrees of Freedom) because of the underactuated systems that have less control inputs than degrees of freedom. To overcome the constraints, a quadrotor with tilting rotors has been proposed. However, there are a loss of thrusts due to the tilt of the rotors and a decrease in maintenance due to additional four servo motors. In this paper, we propose a novel quadrotor with a dual-axis tilting frame using a parallelepiped mechanism. Using its parallelepiped mechanism, it has 6 DoF, that is, it can tilt its frame in the roll and the pitch direction independently by actuating only additional two servo motors. Also, it has same efficiency as the conventional quadrotor since all rotor surfaces are always kept parallel by the dual-axis tilting mechanism. Furthermore, we show a design procedure for the mechanism that achieves the given maximum folding ratio without the rotor overlapping vertically in any transformation state. In flight experiments, we confirm that it is capable of hovering at a desired position and translational motion in a desired circular trajectory with the dual-axis transformation in the roll and the pitch direction.
Virtual reality technology has become widespread and motion platforms have been developed. However, the relationship between motion and sensation in MP has not been fully evaluated. Therefore, in this study, we confirmed that front and back motion and continuous rotation synchronized to the video could affect the human sensation of rotation. Then, we experimentally estimated the threshold of rotational sensation that humans can perceive, and tested the relationship between MP motion and sensation. The experimental results confirmed that 75% of people could not perceive continuous rotation up to 25 [deg/s] and that it may not affect the physical sensation.
Abstract It has been reported that monitoring of older people's safety can improve the quality of their life in nursing homes and reduce the burden on care professionals at nighttime. The aim of this paper is to report the progress in our efforts of developing a bedside robot which functions as an Input/Output device between a monitoring system and older people.
We developed an one board, USB power supply type proximity sensor. The sensor can measure distance (0〜30[mm]) and tilt angles (-45 to +45[degree]) within 1ms. The resolution of distance measurement is less than 26.7[μm], and average error is less than 33.1[μm] in close distances. The sensor is suitable for robot application such as control of fingertip position precisely at high speed. For the control, we describe a method of setting an optimum gain that combines an evaluation function calculation and an asymptotic stability analysis.
We propose a system that estimates the health condition of older adults at home based on the responses in which they talk with a robot. The system actively elicits the clue for estimation of the health condition by not only observing a living environment passively but by talking to older adults. As a first step, to survey how older adults react to the voice-call robot, we conducted a user study in a situation that the robot talks to older adults who were about to leave a chair. In this paper, we report the user study and discuss appropriate talking.
Since 1988, the Department of Electronic Control System Engineering at National Institute of Technology, Numazu College has conducted development of Micro Intelligent Robot System ``MIRS'' as a Project-Based Learning (PBL) into the curriculum for the education of system development. We are intended to provide a place of learning ``process of system development'' and ``making education as a creative education'' and ``Proposal and management of competitions with social implementation'' through the development of robot. In this paper, we report on the results of the socially implemented robot education conducted on the theme of Life with Robots. The results of the three years of practice showed that the teams that conducted social implementation experiments had higher technical evaluation scores than those that did not. However, it was also found that even the teams that did not do social implementation experiments got higher evaluation points at the robot presentation. In other words, social implementation is one component of achieving the objectives through robot development education.
With the recent development of deep learning technology, model learning of low-dimensional potential dynamics embedded in high-dimensional observation data has attracted much attention for complicated robotic control on unknown environment. In previous research, this model learning is formulated as a variational lower bound maximization problem of the marginalized log-likelihood of observed time-series data. From the viewpoint of multi-objective optimization, this problem can be interpreted to be a linearly weighted sum of multiple objective functions, which can only find the convex part of the Pareto front even by optimizing the weight parameters. To find all the Pareto front regardless of its shape, this paper proposes a new model learning framework based on multi-objective optimization with the augmented Tchebyshev scalarization. In the simulation of the object manipulation problem by a robot manipulator with camera images as observation, the effectiveness of the proposed method is demonstrated through Baysian optimization, which efficiently explores one preferred solution on the Pareto front.