Search

A projected virtual hand interface, which visually extends the arms of users using projection images, enables wheelchair users to reach unreachable objects; however, its projection area is usually smaller than the reaching area for its users. In our previous research, we developed a wheelchair system with a projected virtual hand system in which users can move the projected area via finger movements on a touch panel. However, it requires some time to move the projected area to the desired position. To address this problem, we propose a wheelchair system enhanced with a projected virtual hand that allows controlling a projection area using a user's head orientation. The proposed system measures the current user's head orientation and distance between a user and a projection surface. Then, it adapts the suitable pan and tilt of the projector by considering its positional relationship with the projection plane. As users can operate the projection area simply by turning their head, this operation can be executed simultaneously with operating a virtual hand using their hands. We propose a control model for projector rotation according to the user's head orientation. We implemented a prototype based on the model and confirmed the system's latency. The usability study revealed that the proposed method enables users to perform pointing tasks in a shorter time compared with the existing method. Moreover, it has acceptable interface usability.

Human body augmentation makes it possible to obtain new abilities that we cannot achieve with our actual bodies. A projected virtual hand interface is a promising approach for body augmentation because it can extend a user's reach in daily life without the need to wear a device. Although users can manipulate a projected virtual hand as if it were their own hand and can interact with distant objects through it, they cannot feel the sensation of touch when the projected virtual hand is overlaid on a real object. In this paper, we propose a novel pseudo-haptic feedback framework to provide users with the tactile texture of objects without the use of haptic devices. We designed three types of visual effects that produce unevenness, slipperiness, and softness. The experimental results indicate that the proposed visual effects can lead users to feel the intended tactile sensation. Furthermore, the visual effects provide users with tactile sensations with three to five levels of intensity without producing a strange feeling.

Tremors, which refer the trembling or shaking of body parts, are caused by the involuntary contractions of muscles and are one of the most common movement disorders that often reduce individuals' the levels of the hand functions and also affect their independent body movements adversely. ``Extended Hand (EH)”is an augmented body technology, which visually extends the length of a user's arm by a spatial augmented reality. It could facilitate individuals whose reaching areas are limited by body movement disorders to interact with distant objects. However, the technology is currently limited, because any trembling of the hands contacting with the touch panel of the EH system is amplified, causing EH to shake in the physical world as well. To make EH benefit individuals whose hands and body are disabled by tremors, we proposed methods for stabilizing the movements and gestures of the EH to allow EH controlled steadily by their trembling hands. We developed a hand tremor simulator by which trembling hands interacting with EH were imitated, and investigated the effectiveness of stabilization methods by conducting a series of pointing and grasping experiments. The stabilization method composed of Kalman filter and Hysteresis was confirmed to be effective in reducing the unstable translation and unstable stationary state of EH, and the method combining the stabilization of the distances from the fingertips to their centroid and the analysis of the change in low frequency components of the touching areas effectively enable trembling hands to control the EH gestures steadily.

This paper describes the utilization of the downwashes of multicopters for gas-sensing applications. Multirotor drones are an attractive platform for sensing applications. Their high maneuverability enables swift scanning of a target area with onboard sensors. When equipped with a gas sensor and used for gas-sensing applications, however, the strong downwash produced by the rotors poses a problem. When a multicopter is hovering at a low altitude, gas puffs leaked from a gas source on the ground are all blown away. Here, we propose to use two multicopters connected by a rod or a string and place a gas sensor at the midpoint of the rod/string. The downwash generated by each multicopter spreads radially after it impinges on the ground. When two multicopters are connected, the airflows spreading radially along the ground from the two multicopters impinge at the center and are deflected in the upward direction. Gas puffs wafting near the ground surface between the two multicopters are carried by this upward airflow to the gas sensor. Experimental results are presented to show the soundness of the proposed method. The connected quadcopters hovering over an ethanol gas source was able to detect the gas even with a moderate cross-flow.

Here, we report on computational fluid dynamics (CFD) simulations conducted to develop a chemical sample collection device inspired by crayfish. The sensitivity of chemical sensors can be improved when used with a sniffing device. By collecting fluid samples from the surroundings, all solute species are also collected for the sensor. Crayfish generate jet-like water currents for this purpose. Compared to simply sucking water, food smells dissolved in the surrounding water can be more efficiently collected using the inflow induced by the jet discharge because of the smaller decay of the inflow velocity with the distance. Moreover, the angular range of water sample collection can be adjusted by changing the directions of the jet discharge. In our previous work, a chemical sample collection device that mimics the jet discharge of crayfish has been proposed. Here, we report CFD simulations of the flow fields generated by the device. By carefully configuring the simulation setups, we have obtained simulation results in which the angular ranges of the chemical sample collection in real experiments is well reproduced. Although there are still some discrepancies between the simulation and experimental results, such simulations will facilitate the process of designing such devices.

Convolutional Long Short-Term Memory Neural Networks (CNN-LSTM) are a variant of recurrent neural networks (RNN) that can extract spatial features in addition to classifying or making predictions from sequential data. In this paper, we analyzed the use of CNN-LSTM for gas source localization (GSL) in outdoor environments using time series data from a gas sensor network and anemometer. CNN-LSTM is used to estimate the location of a gas source despite the challenges created from inconsistent airflow and gas distribution in outdoor environments. To train CNN-LSTM for GSL, we used temporal data taken from a 5 × 6 metal oxide semiconductor (MOX) gas sensor array, spaced 1.5 m apart, and an anemometer placed in the center of the sensor array in an open area outdoors. The output of the CNN-LSTM is one of thirty cells approximating the location of a gas source. We show that by using CNN-LSTM, we were able to determine the location of a gas source from sequential data. In addition, we compared several artificial neural network (ANN) architectures as well as trained them without wind vector data to estimate the complexity of the task. We found that ANN is a promising prospect for GSL tasks.