-
原稿種別: 表紙
2003 年 8 巻 3 号 p.
Cover1-
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
-
原稿種別: 目次
2003 年 8 巻 3 号 p.
Toc1-
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
-
原稿種別: 目次
2003 年 8 巻 3 号 p.
Toc2-
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
-
藤本 英雄, 佐野 明人
原稿種別: 本文
2003 年 8 巻 3 号 p.
227-
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
-
舟橋 健司, 岩堀 祐之
原稿種別: 本文
2003 年 8 巻 3 号 p.
229-235
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
In this paper, a model to realize an interactive manipulation of virtual liquid using a virtual vessel is described first. Then, we extend this model for the liquid flow. We consider that a liquid vessel is separated into a grid and define the density and the average velocity of liquid in each grid. We decide the movement of a liquid using these value, then we can realize an interactive manipulation of the flowing liquid. Our system with this proposed model makes it possible to catch the liquid, then to flow and spill the liquid by tilting it.
抄録全体を表示
-
植田 直樹, 櫛浜 斎延, 平井 慎一, 田中 弘美
原稿種別: 本文
2003 年 8 巻 3 号 p.
237-246
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
Recently, haptic interface has been intensively studied for providing a sense of touch sensation. Moreover, high-performance force feedback displays have been developed for realizing haptic interface with the virtual environment. In this paper, we propose a novel approach to extraction of rheological properties of deformable objects based on haptic vision approach, which was proposed for vision-based automatic construction of virtual space simulators. We first observe an object by active vision to obtain its 3D shape and posture. We then estimate a contact point and a contact force to cause deformation behavior of the object most stably and effectively, based on the acquired shape and posture. Next, we exert the contact force to the object by active touch, i.e., by Push operation of a robot hand. Rheological properties of the object are estimated and extracted from the observation of the contact force transition and deformation analysis, using a loadcell-sensor and a range finder observing object's behaviors. Experimental results showed the validity and effectiveness of the proposed approach.
抄録全体を表示
-
友國 誠至, 杉山 勇太, 平井 慎一
原稿種別: 本文
2003 年 8 巻 3 号 p.
247-254
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
We describe virtual rheological objects computable in realtime. Deformable soft objects such as food and tissue show both elastic and plastic properties, which are referred to as rheological objects. Construction of virtual rheological objects has not been, however, established yet. In this article, we construct virtual rheological objects, which are applicable to realtime presentation of deformation and force of rheological objects. First, we formulate three-element model to describe the rheological deformation using a truss model. Second, applying an extended penalty method, we maintain the topology of the truss model and describe the interaction among objects. Next, we demonstrate coherent description of rheological objects, elastic objects, and viscoplastic objects. Finally, we demonstrate the realtime computation of virtual rheological objects.
抄録全体を表示
-
宮崎 慎也, 遠藤 守, 山田 雅之, 長谷川 純一, 安田 孝美, 横井 茂樹
原稿種別: 本文
2003 年 8 巻 3 号 p.
255-261
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
A fast computation elastic model for real-time processing, 'gradational element resolution model' was developed, constructed by a small number of elements. This paper realizes the function of cutting objects. Object deconstruction by cutting is fundamentally performed by releasing vertex sharing between elements. When more than a given amount of force is worked to separate a pair of elements, connections between them are released. Elastic element layout and vertex sharing are reconstructed according to shape metamorphosis by cutting interaction. The reconstruction process is limited and influenced by the metamorphosis to avoid increase in motion computation time.
抄録全体を表示
-
田川 和義, 林 宏卓, 木島 竜吾
原稿種別: 本文
2003 年 8 巻 3 号 p.
263-269
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
In spite of many dynamic cloth simulation methods have been proposed in computer graphics and textile engineering field, there are no quantitative evaluation method for comparing real and virtual with various dynamic deformations that can be seen our daily life. Therefore dynamic cloth simulation method have been improved based on subjective evaluations. In this paper, the evaluation experiments in vacuum is proposed and sevral result of experiment is evaluated and discussed.
抄録全体を表示
-
野上 良, 榎 亮, 登尾 啓史
原稿種別: 本文
2003 年 8 巻 3 号 p.
271-278
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
In this paper, we propose three kinds of mass-spring-damper (MSD) models of rheology object, and experimentally evaluate the best one concerning to shape and volume accuracies. The MSD model requires a few cost to calculate force propagation and shape deformation of rheology object. For this reason, the dynamic animation can be made by a personal computer within the video-frame rate (about twenty milli-seconds). Moreover, in order to maintain deformation precision, we calibrate all coefficients of dampers and springs under many experimental data by the randomized algorithm. Then in the set of simple pushing operations, volume and shape of virtual rheology object based on the best model are extremely similar to these of real rheology object. This is a case study to generate dynamic animation efficiently and precisely by the MSD model.
抄録全体を表示
-
前川 佳徳, 河崎 雷太
原稿種別: 本文
2003 年 8 巻 3 号 p.
279-285
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
Virtual buttocks, which can simulate the contact deformation with some object in a real situation, is developed. By the pressure distribution obtained by the simulated contact deformation of the virtual buttocks, for example, with some seat, the comfortability of the seat can be evaluated. The virtual buttocks is modeled as a homogeneous chunk of flesh with a pelvis and a femur, and the homogeneous chunk is modeled as a hyper elastic one (Ogden model). A contact deformation is simulated by a finite element method. Although the simple buttocks model is employed, a simulated deformation of the buttocks for a seating case, the contact figure or the pressure distribution, has a good accuracy to the experimental one, by using of the material characteristic values of Ogden model, which are identified by the experiment.
抄録全体を表示
-
若松 栄史, 加藤 恒範, 妻屋 彰, 白瀬 敬一, 荒井 栄司, 平井 慎一
原稿種別: 本文
2003 年 8 巻 3 号 p.
287-294
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
A modeling of linear object deformation including knotted shape is presented. Deformable linear objects such as tubes, cords, and wires are used widely; data transmission, object transportation, fixing or packing of objects, and so on. Such manipulative tasks include knotting. They can be also knotted in order to make them compact for their storage or transportation. Therefore, it is important for linear object manipulation to analyze knotting. In this paper, we develop an analytical method to model the shape of a deformable linear object, especially its knotted shape. Firstly , a geometric representation to describe the shape of a linear object with flexural and torsional deformation is introduced. The internal energy of the object and the geometric constraints imposed on it are then formulated. The shape of the object in the stable state can be derived by minimizing the internal energy under the geometric constraints. Secondly, procedure to compute the deformed shape is developed by applying a nonlinear programming technique. Thirdly, the above method is applied to the knotted shape of the object. The object contacts with itself at some regions and frictional force yielded there can influence its shape when it is knotted. Therefore, friction by self-contact of the object is considered. Finally, the validity of our method is demonstrated by measuring the knotted shape of rubber cord.
抄録全体を表示
-
井上 健司, 上杉 麗子, 笹間 亮平, 新井 健生, 前 泰志
原稿種別: 本文
2003 年 8 巻 3 号 p.
295-302
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
A new haptic device using flexible sheet such as rubber is proposed. Controlling the bias tension of the sheet, the device varies compliance of the sheet in the normal direction; users can feel compliance of various virtual objects if they touch the sheet. The device is suitable for displaying soft objects. An elastic plane model offlexible sheet is analyzed by FEM. A prototype device using a square rubber sheet is developed. This device controls the bias tension of the sheet by pulling its four corners with four motors. The normal force applied to the center of the sheet and the normal displacement at this point are measured, and the obtained force-displacement curves of three kinds of sheet-urethane, chloroprene rubber and low elastic rubber-are compared. The results show that the developed device can generate different compliance by changing the bias tension. The compliance of human palm is measured and simulated by the device.
抄録全体を表示
-
福岡 久雄, 佐藤 文明, 坂下 善彦, 水野 忠則, 和田 雄次
原稿種別: 本文
2003 年 8 巻 3 号 p.
303-310
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
This paper discusses a model for event distribution in which several distributed virtual environment systems, provided by Internet service providers (ISP), can be interoperable, and analyzes its dynamic characteristics. Each ISP has a server that manages a virtual world called scene. The interoperability is realized by connecting these servers. Each client has a connection to a predetermined server. When a client is connected to a server and its user belongs to some other server's scene, the former server plays a role of proxy between the client and the latter server. We have clarified the processing mechanism of servers for event distribution and have analyzed its dynamic characteristics using the queuing system theory. As a result, we have obtained an equation to show the limitation ofserver's scalability in terms of each user's event occurrence rate. The validity of the equation has been confirmed by the comparison to a computer simulation result.
抄録全体を表示
-
河野 武, 伴 好弘, 上原 邦昭
原稿種別: 本文
2003 年 8 巻 3 号 p.
311-319
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
This paper proposes a new kind of visual marker for the wearable computer based on Augmented Reality(AR) system. In the wearable AR system, the markers are often used as position tracking subsystems. But, these subsystems have some disadvantages such as the visual markers' small amount of data types, small recogniton area and so on. The visual maker we developed has two features. First, it has a large amount of marker types. The marker consists of 32 bits length of data area with error detection code. Second, it is a two stage marker recognition. When the marker is far away from the system, it is used as a position tracking mark. It includes a code recognition along with the position tracking when the marker is close to the system. These features enable a wearable AR system to have various utilities.
抄録全体を表示
-
今西 勁峰, 中尾 恵, 黒田 知宏, 小山 博史
原稿種別: 本文
2003 年 8 巻 3 号 p.
321-328
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
Conventional display for robotic surgery such as flat displays and stereoscopic displays decreases obtainable information around target tissue. This paper proposes a haptic navigation method which draws master manipulator back from untouchable regions around target tissue by giving repulsion on master manipulator. This study developed an input interface for designation of 3D untouchable regions through 2D device, and generation method of stable repulsive force. Simulator based experiment clears effectiveness of proposed force navigation method for improving safety of robotic surgery.
抄録全体を表示
-
松下 伸行, 勝部 元, 日原 大輔, 後 輝行, 兼松 幸市, 吉村 真一, 暦本 純一, 山本 喜一
原稿種別: 本文
2003 年 8 巻 3 号 p.
329-338
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
This paper describes our design and implementation of a new augmented live stage system called Motion Groove. Our system is a new application of a Mixed Reality system. On a live stage, because of lighting production, the optical environment changes violently. Conventional positioning systems used for motion capture cannot be used in this environment. We have developed a positioning system that can be used during live performances on stage. The positioning system has three features. The system can recognize multiple optical beacons simultaneously up to 255. It can track optical beacons that move quickly such as when attached to the limbs of dancers. It is robust to changes in the optical environment e.g. complete darkness, spotlights, sunlight. We implemented the live stage augmentation system called Motion Groove using a positioning system, and confirmed its practicality.
抄録全体を表示
-
梶本 裕之, 稲見 昌彦, 川上 直樹, 舘 [ススム]
原稿種別: 本文
2003 年 8 巻 3 号 p.
339-348
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
This paper introduces a new type of Haptic AR system. Our approach uses electrical stimulation through the skin, or electro-tactile display. Since the electrotactile display essentially requires only electrodes on the skin, it can be made quite thin and small. This leads to a new application area of tactile display, the SmartTouch. The SmartTouch is composed of a thin electro-tactile display and a sensor mounted on the skin. The sensed information is converted to tactile sensation through electrical stimulation. Thus, the wearer not only can make physical contact with the object, but also "touch" surface information of any modality, even those that are ordinarily non-touchable. The prototype of SmartTouch is composed of two layers. The top layer has a 4 by 4 array of stimulating electrodes on the front side of a thin plate, while the bottom layer has optical sensors on the reverse side of the plate. Visual images captured by the sensor are translated into tactile information, and displayed through electrical stimulation. We endeavored to realize the perception of luminance information as the unevenness of the object surface, by imitating the sensory nerve activity with electrical pulses.
抄録全体を表示
-
則武 厚, 渡邊 洋, 梅村 浩之, 松岡 克典, 八木 昭宏
原稿種別: 本文
2003 年 8 巻 3 号 p.
349-356
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
Those who aimed to create highly dynamic and interactive virtual reality (VR) spaces assumed that observers accurately localized where stimuli were presented. However, we don't know so much about how accurate people localize the stimuli in such created VR spaces. To investigate the localization, we studied the representational momentum (RM) phenomenon, which was the mislocalization of disappearance positions of the moving target, in VR space. The reasons for dealing with this phenomenon are 1) that this phenomenon could be referred to as the basic component of highly dynamic and interactive VR applications and 2) that because many studies reported this phenomenon in two-dimensional space, we could compare our results with those of the previous studies. This experiment was designed to confirm RM phenomenon not only when a target was presented but also when a static reference was presented with a target. The mislocalizations of the moving target along the horizontal axis were measured in the horizontal and the depth axes. The mislocalizations were observed in the direction of the moving target in the horizontal axis but not observed in the depth axis when the target was approaching to the front of the subjects, whether the reference was presented or not. And the mislocalizations were not observed in both horizontal and depth axes when the target moved away from the front of the subjects. These results suggest that in VR spaces, people would localize the moving target in more forward than the actual position when the target approaches to the subjects' midline. These results also suggest that these biases would be made under VR applications in which perceptions of moving targets interact with other perceptual modalities.
抄録全体を表示
-
原稿種別: 付録等
2003 年 8 巻 3 号 p.
357-
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー
-
原稿種別: 表紙
2003 年 8 巻 3 号 p.
Cover2-
発行日: 2003/09/30
公開日: 2017/02/01
ジャーナル
フリー