In laparoscopic surgery, surgeons cannot use large and complicated shaped traditional instruments for abdominal surgery and their hands directly. When it is difficult to perform surgery by using only instruments for laparoscopic surgery, surgeons often make a 7-8 cm incision through which large instruments or their hand can be inserted. This paper propose an assemblable mechanical hand like a human hand that can be inserted through trocars. We developed an assemblable three fingerd five-DOF hand. To drive its fingers, power is transmitted with wires, gears and shafts. Experimental results verify that the hand can grasp and pushing aside various objects.

View full abstract

This paper described the small-sized omni-directional 3D laser range scanner. This sensor can scan surrounding 3D form, because the existing 2D laser range scanner is rotated in the direction of a yaw angle by a turntable. Since the turntable is made by a steppingmoter and a slipring, a highly accurate positioning is possible. The accuracy of the sensor is yaw angle resolution less than 0.5 degrees and measurement more than 300,000 points. By modeling, a surface model is made using the point data obtained from this sensor. The modeling is to do connection of triangle patch and calculation of normal vector. Formerly, a target object can not be extracted from obtained data. However, extraction of the target object is possible according to background difference.

View full abstract

This paper deals with a velocity measurement method for robotics. The authors have proposed "synchronous-measurement method (S method)" that measures the velocity in sync with alteration of pulse numbers of an optical encoder. The method has an advantage that accurate velocity measurement is achieved in all speed ranges. The method, however, has a technical issue in practice: accuracy often deteriorates with nonideality of interpulse angles in the optical encoder. Only complicated algorithm with conditional branching can solve this issue. Hence this study proposes a method that solves the issue by a simple algorithm without any branching. Simulation results show the measurement accuracy while experimental results verify control performance of the proposed method.

View full abstract

Biped walking causes translational acceleration and vibration on the torso where an attitude sensor system is usually implemented. The accuracy of measurement tends to be low because of those bad influence. We propose a method of attitude estimation system that takes translational acceleration into account. Planned translational acceleration is derived from generated walking pattern, and subtracted from the output of the accelerometers, so that the effect of acceleration caused by robot motion is reduced when the direction of gravity acceleration is estimated. An attitude measuring system which consists of 3 fiber optic gyroscope sensors and 3 servo accelerometors was developed. The system was attached on a full size humanoid HRP-2, and the proposed method was implemented. The performance of the proposed method was evaluated during normal walking of HRP-2.

View full abstract

In this paper, we propose a new coded structured light projection method that aims at three dimensional image acquisition at a high speed frame rate. The proposed method can select not only temporal encoding but also spatial encoding according to image features depending on brightness change. We construct a verification system that is composed of a DLP projector and an high-speed camera, and we evaluate the effectiveness of the proposed method.

View full abstract

This paper presents a method for detecting and tracking multiple moving objects with multiple 2D laser range sensors(LRS's). The coordinate systems of multiple LRS's are calibrated based on Kalman filter and Nearest Neighboring Standard Filter(NNSF). The moving objects are tracked based on Kalman filter, Covariance Intersection(CI) and the assignment algorithm. A rule based track management system is embedded into the tracking system in order to improve the tracking performance. The experimental result in indoor environments validates the proposed method.

View full abstract

This paper proposes a method for multiple persons tracking using sensing floor and multiple cameras. Our method estimates 3D position and ID of targets. To integrate sensing floor and multiple cameras, our method uses particle filter framework. The particle filter framework is known to be effective for tracking people. In this framework, our method adopts a load on sensing floor, color histogram and background subtraction of cameras images for hypothesis evaluation. Our experimental results demonstrate the effectiveness and robustness of the method against complicated movements of multiple persons.

View full abstract

This paper proposes a construction method of dense ray space using depth components. Light Field Rendering is the technology that can compose a free viewpoint image. However, there is a weak point in Light Field Rendering. A high-resolution free viewpoint image needs dense ray space. This construction method supplements a weak point of Light Field Rendering that many cameras are necessary. This construction method composes a ray of a spot without a camera. Therefore, this construction method doesn't need many cameras. Lastly, this paper considers a construction method of dense ray space using depth components.

View full abstract

Recently, it is noticed that the Computer-Based Testing is getting a lot of attention as personal computers become popular and the internet society is growing. At present, CBT is normally used to measure examinees' knowledge or intelligence. But, it is not considered enough to measure examinees' skill for now. In this paper, we propose a method to measure examinees' skill taking putt swing for example. The task is easy to analyze and compare because the initial condition of the task can be fixed easily. We got motion data using a motion capture system. And we analyzed the motion data.

View full abstract

Recently, computer-based testing that automates examination using computer is attracting a lot of attention due to the development of information technology and of the spread of computer network. This paper describes a method to automate a test in which examinees actually perform the task to solve given problems, taking putt swing for example. Our testing system assesses the performance of examinees from motion data. We got motion data using a motion capture system. In this paper, we propose a method using recurrent neural network to estimate the expertise of examinees.

View full abstract

An infant robot has many degrees of freedom. By making constraints and coordinating many DOF by using "intensions" such as "want to stay sitting," or "want to move more forward," etc., various motion patterns emerge according to a particular intension. We traced acquisition process of infant locomotion by changing the intension according to the process of development through numerical simulations, and obtained autonomous locomotion patterns.

View full abstract

In this paper, methods of detecting planar regions by LMedS estimation using a relative disparity image are proposed and discussed. LMedS estimation estimates proper models for a relative disparity image and outliers are regarded as obstacles. Fundamental experiments show that the proposed methods are effective for detecting planar regions when a humanoid is not walking. Then it is verified that the humanoid can detect planar regions by the proposed methods while walking.

View full abstract

A method of range image sensing by integration of active and passive methods using multiple spotlights is proposed. Using the information of active stereo sensing makes the measurement time of passive stereo sensing by a stereo camera shorter and makes its accuracy higher.

View full abstract

Our research objective is construction of 3D map and development of 3D control interface. It is necessary to measure correct 3D environment shape. We had developed 3D scanners for the measurement. In these researches, 3D shape was measured by rotating a 2D laser range finder. These scanners could measure correct 3D shape in static environment. However, these scanners did not measure correct 3D shape when there were moving objects. Moreover, density of 3D shape was not uniform. In this paper, the authors will propose new scanning method which can solve these problems. We will show a design of this new scanner.

View full abstract

Our research objective is construction of 3D map and development of its practical applications. As an application, the authors had proposed 3D control interface for an operator who controls remote control robot. It is well know that an operator can control a robot easily using bird's-eye view. Using 3D control interface, an operator can control the robot at bird's-eye view. The bird's-eye view is constructed from 3D map and robot model. The operator can recognize relationship between the robot and the environment from 3D map and robot model following real robot motion. In this paper, we will describe 3D control interface and its implementation for HELIOS carrier. We will show an experimental result of 3D control interface. From result, we confirmed the effectiveness of the proposed method.

View full abstract

Time is crucial in applications such as sensor data fusion, autonomous mobility and SLAM. However clocks at end systems are rarely synchronized and often running at different speeds, therefore this lack of synchronization reduces the accuracy of sensor readings. The new SCIP2.0 protocol of Sokuiki sensor allows acquiring time values by timestamping range readings. Our work consists in a method for time synchronization with clock skew estimation, between a Sokuiki sensor and a host computer.

View full abstract

Tracking humans' position is a useful skill for the coming generation of mobile robot. It is a challenging problem of planning and control in dynamic environment. We propose the omni-directional estimation method of speaker's position using the combination of audio and visual information. Estimation of the position of the sound is carried out to calculate the difference of arrival time from sound source to multi-channel microphones. The robust human template matching on the omni-directional image is employed to combine the result of sound source estimation to realize a highly accurate estimation of speaker's location. In our experiments, the systems were implemented and tested on an omni-directional robot at our laboratory. The results show that we are able to reliably detect and track moving objects in natural environment.

View full abstract

This paper describes road sign localization using a single camera and 3D environmental model. The proposed Mobile Mapping System (MMS), featuring the carrier phase D-GPS/DR (Dead Reckoning) combined navigation system and the GPS-Gyro/IMU(Inertial Measurement Unit), performs highly accurate positioning performance in centimeters and posture estimation at 0.073[deg](1σ) for heading, 0.064[deg](1σ) for pitch and 0.116[deg](1σ) for roll. In road sign measurement environment, there are many kinds of objects near a road sign which cause serious misidentification. The proposed localization method utilizes different LOS images to identify collect road sign's corresponding laser data to improve localization accuracy.

View full abstract

Toward better understandings of emotions, we argue functions of emotions from embodiment and phenomenological perspectives. Based on the phenomenological assumption that emotions are actions to reduce "dissonance", a synthetic model is proposed. As an experiment, we examine how spatio-temporal dilemma is reduced, an instance of which is that if we eat too much because we are hungry, then, consequently we vomit. We show that our model can reduce the dilemma by exploreing and exploiting an agent-environment dynamics in the emergent manner.

View full abstract

Humans can execute various motor tasks in daily life, sports, and so on. Since human body as musculoskeletal system has multiple degrees of freedom, the notion of motor primitive is one of the keys for human behavior generation and motor control. One of the acquisition strategies of motor primitives is that action units as motor primitives are extracted from transitional movements in physical structure of body and environment. Our purpose is extraction of motor primitives from interaction between a body and a environment. We considered about baby motor primitive acquisition. Motor commands which are random walk with a drift generate for searching a body and environment dynamics. The motor command is sufficient to generate behavior for baby, and a physical information structure appears as frequency distribution of baby's head direction. In conclusion, motor primitives are successfully extracted as transitional movement between one stable area to another area.

View full abstract

We have meetings for the discussion on the function and interface of next generation area sensor (SOKUIKI sensor). The purpose of these meetings is to discuss on how to use the area sensor for service robots and expected specifications of the next generation sensor. The basic designs of the command interface for such a kind of sensor are also discussed. This paper reports the activity of our group.

View full abstract

Recently, SOKUIKI sensor is used in various fields including robotics. SOKUIKI sensor "URG-04LX" can take distances by measurement. And by using special mode of the firmware, Received Light Intensity and AGC Voltage Level can be obtained. This paper describes a method to estimate Gray Level of scanned objects by using Received Light Intensity, and the method uses AGC Voltage Level to accurate estimation.

View full abstract

This paper describes a real-time system for measuring the three-dimensional shape of solder bumps arrayed on an LSI chip-size-package (CSP) board presented for inspection based on the shape-from-focus technique. It used a copper-alloy mirror defonned by a piezoelectric actuator as a varifocal mirror to build a simple, fast, precise and silent focusing-mechanism. Accurate measurements of bump height were achieved with errors of not more than 10 μm. This met the requirements for measuring coplanarity. A practical measuring speed of 1.7 s/chip for a small CSP (4 x 4 mm^2) was achieved by incorporating an exclusive FPGA processor to calculate focus measure and by building a domed array of LEDs as an illumination system with high intensity and uniformity so that a fast (150 fps) and high-resolution (1024 x 1024 pixels/frame) CMOS image sensor could be used.

View full abstract

We are developing the exfoliative tile and crack inspection robot in wall, in order to prevent beforehand collapse of a old building wall tile. The system is incorporated to the wall surface operation robot,which can detect the exfoliative tile and detect the crack more than 0.3 mm in wall tile. In this paper the validity of the system for wall tile inspection is described

View full abstract

A method of scooping rock piles automatically by wheel loaders is proposed based on reaction forces from rock piles. First, reaction forces are analyzed by some fundamental experiments using a miniature wheel loader system with the same bucket driving mechanism as an actual wheel loader. Using this result, the model of reaction forces against rock piles in scooping operation are constructed.

View full abstract

This paper describes a method for estimation of pose (position and orientation) of large objects such as construction materials using multiple ID devices attached to the objects. Object-pose tracking is crucial issue for integration of the objects and their information via ID devices since the workers or robots that do not have ID reader can change the object pose. The method uses the position and orientation of the ID reader with respect to the reference coordination frame. This paper focuses on the verification of the estimation result based on data acquisition through ID devices. The pose estimation method are not stable when the pose error of the ID reader on acquisition of the IDs. The verification method uses only the poses of the ID reader when the reader obtains the ID of the devices, and the estimated object pose. Experimental results show feasibility of the proposed method.

View full abstract

We propose a robotic system to assist workers while dismantling office interiors, due to these tasks demand a great physical effort. Particularly, this report shows the steps for positioning automatically the robot's tool on the items to be removed and also it includes a suitable teleoperation process. The system consists of an articulated arm mounted on a movable platform, a desktop device for the robot arm teleoperation process, an stereo vision system as main sensor, and a laser pointer as a guidance mark. The last two devices are attached on the robot.

View full abstract

New robot system will be playing an essential role in the future dismantling service for renewal of office in-fill facilities in buildings. In our preliminary study the dismantling task would be achieved by proper collaboration of human workers and robots utilizing mobile arms, interface devices and software. This report discusses the system concept, the system requirement, how robots and workers would collaborate.

View full abstract

Contour following by using hydro dynamical construction machinery, such as hydraulic excavator for example, were said to be difficult. The purpose of this research is to develop new passive compliance built-in frontal part of the machinery for easier contour following. In this paper, mechanical model of the boom with elastic element was built, and direct stiffness method was applied to analyze the compliance of the frontal part. The compliance characteristic of the front part of the boom was achieved with this simulation result. It was compared with the experiment result. The result of both was almost corresponding, and the validity of an analysis method was shown.

View full abstract

An imaging system with a focusing mechanism based on perfect perspective projection was constructed using a movable image sensor to achieve high-quality three-dimensional (3-D) imaging and precise measurement of shapes in close range based on the shape from focus technique. The image sensor was mounted on a single-axis movable table. An aperture stop was located at the object focal point of a fixed image-taking lens. Magnification was proportional to image sensor displacement. This was used to obtain the transversal coordinates of a measurement point. An object was rotated around a fixed axis to be observed by the imaging system from all directions. A method of correctly identifying the rotation axis was presented and ensured that range data obtained at different observation angles could be combined precisely at a standard deviation of 10 μm. The whole shape of a mechanical part such as a screw was measured successfully.

View full abstract

This paper proposes a high frame rate 3D imaging system based on structural light projection using temporally-coded binary patterns. This system consists of a DMD (Digital Micromirror Device) projector and a high-speed camera, both of which operate at over 6,000 frames/s. Introducing image processing algorithms of detecting and tracking the projected light patterns enables frame-by-frame 3D image measurement. Experimental results show that the proposed algorithms worked effectively, and the system achieved 3D imaging at 3,350 frames/s.

View full abstract

It is useful for mobile robot navigation or mapping to use three-dimensional environment information. To obtain such information by sensor, one of conventional methods is to rotate two-dimensional laser-range-finder. The method has advantages of (1) wide angle of visibility and (2) small effect of light quantity. However, it has disadvantages of (3) slow-scan speed and (4) existence of limit point. To improve these disadvantages, our research group proposes a novel scanning method. One of the features of the method is that a scanning plane (of the two-dimensional range sensor) is not perpendicular to revolving plane of the sensor. To verify the validity of the method, we performed a basic experiment using a three-dimensional range sensor whose scanning angle can be changed. In this paper, we introduce the scanning method, and report the basic experiment in real world.

View full abstract

This paper gives functional evaluation of the water hydraulic system using ICN (Intelligent Connector) for ICN network of the water hydraulic robot. Proposed system is possible to maintain and change, putt on and take off of machine parts easily. NPA (Network Plug-in Actuator) integrates a motor, a driver, reduction gear, and sensor. NPA realizes easily to install the new parts to a robot, because ICN is succeed in support ICN protocol for read time control. ICN protocol concerned with machinery the only address for indemnifying function and its data. Various water hydraulic machine devices are controlled by ICN and NPA system.

View full abstract

In this research, it is with a water pressure system instead of an oil pressure system. It is going to control a machine by water pressure system. We paid our attention to the energy-saving which water has. We rationalize system management and aim at energy saving.

View full abstract

It is needed to think about ecological condition when we build the building and also dismantle them. This paper proposes new system of categorizing wastes on dismantling for renewal building. RFID tag is useful to categorize wastes effectively based on information of materials or shapes and generate task planning strategy. Developing robot is planed using water hydraulic system with energy saving control methodology based on mini-excavator mechanism. Human and robot cooperate under man-machine control system to reduce the number of workers on removing walls, ceilings, floors or some electric devices in the building and also categorize wastes effectively.

View full abstract

In recent years, unmanned operation systems for the excavator are required in order to realize safety operation under bad environmental condition for human operator and some systems are already developed. However, it is difficult to realize effective operation utilizing such systems. In this paper, we described to examine to extract operator's skill for controlling unmanned hydraulic excavator. We are comparing the operation of skillful operator with non-skillful operator and discuss about the result for modeling of operator's skill. Since these results, we have investigated that skillful operator is tracing uniform excavation trajectory with quick moving and it causes the efficient operation.

View full abstract

This paper describes a detection method of two types of separations for outward tile wall. One of the tile separations is an outer exfoliation separated the tile surface from mortar concrete, the other is an inner exfoliation deteriorated space between the substrate and mortar concrete. In order to research these separations, we focused attention on the wavelet analysis that enable to analyze the frequency element of a waveform on time phase continuously. An impulse sound by a small hammer was measured using a mockup wall that reproduced two types of separations, and then these sound data were inspected by wavelet analysis. As a result, the characteristics of two separations were semi-empirically recognized visualizing the wavelet plane and the quantitative parameters judged the two separations were obtained processing the wavelet data statistically.

View full abstract

This paper describes the adaptive roof structure applying VGT (Variable Geometry Truss). In the study from part 5 to 6, the study of the variable structure applied by the cantilever type was done and the large scale movable monument held on Aichi EXPO 2005 was developed. In part 7, the arch structure roof by VGT and its variable motions were examined. In the arch structure, the both edges of the structure are supported at each hinge. It is very difficult to determine the shape of the structure because these tips are absolutely suited at the hinge position. Using inverse analysis method for shape change of arch structure, the structural characteristics in edge conditions were simulated by dynamical FEM methods. In case of fixed edge, the design of VGT mechanism accompanied with arch motions was enabled. In order to develop the movable roof structure, mechanism of outward panel accompanied with structure motion was introduced in detail. As results of these studies, the optimal and rational shape design of the arch structure was concretely produced.

View full abstract

We developed a new simulator of operation skill training for Double-Front construction Machinery. As for a past construction machinery simulator, the one developed for a lot of educational for dangerous acknowledgment, and to train the operation skill is few. It was simple digging work by the backhoe, and work of the crane. In this research, Double-Front construction machinery which we make a target increases the number of fronts should be manipulated and demands an advanced operation skill. Moreover, the discretion dismantlement and the disaster relief and restoration are developed as assumption work as for Double-Front construction machinery. It is understood that the meaning of the developed simulator is high from these two points. As a result of the proof experiment, we obtained data with a different in the feature transition of expert and beginner's end-effecter orbits. Because the difference was confirmed, it becomes useful as a training machine for the skill improvement by being analyze this in the future and feeding back to the operator appropriately.

View full abstract

In this paper, an edge mode probe sensor using a piezoelectric thin film is introduced. The edge mode is an oscillating mode which the vibration concentrates on the end of a rectangular plate. This probe sensor is used as a probe sensor of the scanning probe microscope. The edge mode probe was oscillated by the PZT thin film deposited on titanium substrate. The PZT film was also used for the detection of the edge mode vibration. The probe had dimensions of 10x2.5x0.1mm. We have succeeded in oscillating and detecting the edge mode vibration by the piezoelectric film. We have also succeeded in detecting the contact by this edge mode probe sensor.

View full abstract

The pneumatic micro dispenser system without any dead volume of sample was proposed. The sampling chip and the 10-directional switching valve chip as the component of this dispenser were developed successfully. The sampling chip succeeded to divide liquid every 1μL or 0.5μL without any dead volume of sample only by the decompression and the pressurization operation of the air, due to the novel detecting method of micro fluid and check valves of hat shape. The switching valve chip selects the outlet that injects the dispensed sample. This chip succeeded to switch ten outlets by built-in solenoid (1N,6VDC) under high leakage pressure (≧500kPa).

View full abstract

We propose a novel type of optically controlled micromanipulation tools that can work in sealed micro space. The micromanipulator arm has a handle to rotate the arm on its long axis. When a focused laser beam is focused on the handle and scanned linearly in the focal plane, the handle is inclined toward the focus. As a result, the arm is rotated on its long axis. The rolling mechanism allows us to manipulate a microobject stably. The micromanipulators with the rolling mechanism were fabricated by using high-speed two-photon microstereolithography. We demonstrated that the manipulators could catch a microparticle in three-dimensions. In addition, a single manipulator arm was successfully used for sliding a microparticle on a cover glass. Such optically controlled micromanipulators offer versatile and powerful tools for micro total analysis systems for biological applications.

View full abstract

This paper presents an optical dynamic gain equalizer (hereafter DGEQ) using an electrostatically controlled linear inchworm actuator (hereafter ECLIA). The actuator consists of silicon sliders, a glass driving stage, a piezo actuator, and a base flame. A slider of ECLIA attenuates an intensity of an optical beam, so that parallel motions of arrayed sliders are able to control and equalize spectrums of demultiplexed signals. A designed DGEQ takes the following advantages of ECLIA to provide gain flattening of spectral gain non-uniformities: Precise motion, Large stroke, Parallel motions of arrayed sliders. The developed feedback control system shows that a single slider position can be controlled within about ±0.7μm from objective position.

View full abstract

This paper reports nerve interface devices for both stimulating and recording by using MEMS technology. The nerve stimulation technology such as a functional electrical stimulation has been studied in order to recover paralyzed function. Our study is interested in a selective nerve stimulating for low invasive FES. This paper presents a literature of our developed interface micro devices for stimulating. We employ micro flexible cuff electrodes for nerve stimulation. In addition, a pneumatic balloon actuator for reliable contact of interface is also discussed. Evaluation of developed devices through animal experiments shows a feasibility of developed devices and requirements for further improvement.

View full abstract

In this study, the bilateral type pneumatic wearable master-slave training device is developed in order to feed back a force information to the trainer. In this paper, the structure of the developed device is discussed, and then the validity of the proposed device is evaluated from the experiments that assume rehabilitation of a knee.

View full abstract

In this study, the assist device of standing up motion has been developed to assist an independent life of elderly or physically handicapped person. The developed assist device consists of a pneumatic rubber artificial muscle and a orthosis. In this paper, after describing a structure of the assist device and control system for knee and ankle joint, the effectiveness for assistance of standing up motion is verified through some experiments.

View full abstract

This paper proposes a new concept of an upper-limb power-assist exoskeleton as a second stage of the research on power-assist exoskeleton robot systems in order to help the activities of physically weak persons. The proposed exoskeleton assists not only the motion of the user but also the perception of the user using sensors. So that the assisted user's motion can be modified based on the environment information obtained by the sensors. The effectiveness of the proposed concept has been evaluated by experiment.

View full abstract

"Muscle suit" that provides muscular support for the paralyzed or those otherwise unable to move unaided is being developed as a wearable robot. The muscle suit consists of a mechanical armor-type frame and McKibben artificial muscle. Using a new link mechanism for the shoulder joints which consist of two half-circle links with four universal joints in total mounted at both ends of each link, all motion for the upper limb has been realized. Application of the muscle suit to elderly people and/or disabled one requires very high-level safety and usability, from the practical point of view. We then in the first place for the practical use, apply the muscle suit to the manual worker. In this paper, we introduce how we can use the muscle suit for the manual worker.

View full abstract

The force sensor with quartz crystal resonator has the characteristic that the resonant frequency changes by the external force, which achieves high sensitivity, high-speed response, and wide measurement range by a simple structure. Also it has the superior feature in the temperature and frequency stability. On the other hand, the quartz crystal resonator had been hardly applied to a dynamic force measurement because of low degree of mechanical characteristic, that is, it is weak to the impact, the bend, and tension. The objective of this study is to construct the sensor mechanism that safely maintains the quartz crystal resonator for the external force with flat structure. We designed and analyzed a novel retention mechanism of the quartz crystal resonator for the flat structure. The proposed force sensor is flat, small, and sensitive enough to be applied to several usages such as medical treatment and dynamic contact force detection of human.

View full abstract

In human skin, four kinds of tactile receptors (Meissner corpuscle, Merkel cell, Ruffini ending, and Pacinian corpuscle) are distributed three dimensionally, i.e., not only on the skin surface but also in the four corresponding depths of the skin. The final goal of this study is achieving a flexible tactile sensor having four layers imitating human skin. An array type tactile sensor made of PDMS, which is a kind of silicone rubber, is developed by MEMS technology. By stacking this sensor one by one, fabrication of a multilayer type sensor is possible in future. In this report, the concept of this sensor is proposed, an array sensor of one layer is fabricated, and the performance of one capacitive force sensing element is characterized preliminarily.

View full abstract