We propose the new concept tools applicable for the rescue use and developed the unified pneumatic pillars. The unified pneumatic pillar can lift up rubbles and make spaces for inserting rescue tools by rescue team members. In addition, they can be used by combining to adapt to the situation. In this paper, we report about improving for practical applications.

View full abstract

A new type of wall climbing and ceiling mobile robot is proposed in this paper. To generate large enough adhesive force even on the rugged wall, a gel mat made of urethane is introduce, whose force can be recovered by the wash. Developed sticky robot based on the above method could realize the 3D mobile performance on the rugged wall and ceiling with the function of washing of the gel mat and monitoring the scene of the inside of the building. The validity of the application to the rescue inspector is also verified.

View full abstract

This research describes the development of a search and rescue mobile robot "Bari-bari-IV", which jacks-up heavy debris while sneaking into a collapsed house. Previous prototypes of the "Bari-bari" series presented some problems, such as traversing uneven terrains, vertical wobbling motion of the debris when sneaking through them, and limited jack-up driving power. Bari-bari-IV is composed of two sets of plates, which are driven in opposite phases in such a way to keep constantly the maximum jack-up height at an overlapping point. In order to realize this method by using a single hydraulic motor, a novel cam is proposed to drive the body back-and-forth and to maintain jack-up height. With this structure, Bari-bari-IV aims sneaking through debris without vertical wobbling motion, traversing uneven terrains and the possibility of jack-up a 1 ton payload. Finally, the effectiveness of Bari-bari-IV is verified by field tests in an environment similar to real situation.

View full abstract

In most case rescue robots are expected to search survivors by moving over or through rubbishing environment. However, the inspection through the rubbish is a complex task, since the passage is narrow and rugged. This paper proposes a new method to aid the locomotion through narrow and rugged space by building a passageway. An inspector is inserted into the narrow space, and it builds a spiral tunnel, which enable carrying the necessary tube. The necessary mechanisms are analyzed for the constitution of a spiral tunnel. The authors have built a prototype of such inspector, and experiments showed the basic motion of the inspector building the spiral tunnel.

View full abstract

This research describes the proposal of a fluid powered ropeway aiming fast search for survivors in half-collapsed buildings. The previous robot "the throw & collect type robot", which was proposed in our lab, consisted of the following sequence : after throwing a parent robot into a disaster area, it throws a child machine equipped with a microphone and a camera, and the inspection is performed while the child machine is retrieved. However, there are two major drawbacks. One is the friction between the tube and debris when the parent robot retrieves the child machine. And the other is the high impact when the child machine falls on the ground. To solve these two problems, we propose a new process "a fluid powered ropeway". After throwing the parent robot, a ball with a flat tube, which is a tube pressurized with high temperature, is shot from the parent robot. And then, a new inspector robot equipped the microphone and the camera moves with the tube using fluid power. The inspector robot capable of both forward locomotion and transverse locomotion is designed to move smoothly with the rube.

View full abstract

This paper describes the connected two unit crawlers to realize various configurations. By changing the configuration of two connected vehicle units in relative positions, the robot with this mechanism can automatically adapt to the surface obstacles in the field, including such complicated structures like debris after disasters. In addition we analyzed the effect of the each axis arrangement in order to realized switching function to compose the four basic configurations more easily, and developed an actual prototype model.

View full abstract

We describe the VoIP system for rescue activity in underground space. The system works on the hybrid ad-hoc network using both wired LAN and wireless LAN, and it enhances the communication distance of voice communication. We realized the communication features and redundancy of prototype system.

View full abstract

This paper is a thing about the development of a robot laying a LAN cable and access point for ad hoc network construction in the closedown space. Electric waves of an underground shopping center or a building are hard to arrive, therefore, closedown space is the space that wireless communication has difficulty. It is necessary to fix the environment that can perform wireless communication so that a rescue robot can perform exploration activity easily in this space. Therefore I build the ad hoc network by wired LAN and wireless LAN.

View full abstract

This paper describe a method for avoiding obstacles using SOKUIKI sensors. Free space is defined by data of SOKUIKI sensor. Positions of obstacle are detected by shape of the free space. Velocity of obstacle is estimated from the positions of obstacle. This method can also estimate velocity of walking people. By the shape of free space and the velocity of obstacle, we define evaluation values that reflect safety for each direction. Then, we decide the target direction and target speed from the evaluation values. Robot can chose safety route and safety speed.

View full abstract

Under the conditions of combustible gas, the rescue operations might be much delayed because the engine cutters that the sparks are generated can not be used as rescue tool. On the other side, as a high pressure water jet is able to cut the obstacles for rescue maintaining the explosion-proof and lower vibration, it is extremely suitable under those conditions, and such a useful development is expected. In this study, Water Jet Cutter System that can be operated at high speed cut under combustible gas has been developed and the practical use of this system has been examined. The system is composed of the field vehicle robot and the water jet cutter. When the disaster takes place, the vehicle robot approaches the rescue site while getting over rubbles, and the water jet cut off the collapsing material. This report describes the evaluations of operating performance for the field vehicle robot and perforating and cutting performance for the water jet cutter to construction materials.

View full abstract

In the past decade, search type rescue robots have been focused on. We have proposed a new mobile mechanism : flexible mono-tread mobile track (FMT), and developed a prototype "Rescue mobile track No.2 (RT-02) WORMY". FMT flexes in shape flexibly in three dimensions when the vehicle turns, climbs up and down stairs. However, it is difficult to measure the posture of FMT using usual sensors. The purpose of the paper is to develop a Flexible Displacement Sensor using nylon string coated with carbon and measure the posture of FMT using Flexible Displacement Sensor.

View full abstract

We developed new rescue robot system that plural rescue robots are remotely controlled by one operator. In this system, a pioneer type robot put multi hop wireless network devices to extend area that surveyor type robots are able to search victims and surroundings in a damaged building. In this paper, evaluation of GUI interface to operate plural rescue robots and distributing system of wireless network device are shown.

View full abstract

To remove toles and pebbles, strong power is needed on the disaster ground. The hydraulic equipments are useful such time. An oil pressure pump is needed to drive the hydraulic equipment. In general, the engine type oil pressure pump is used. Then, we developed the converter that generated the oil pressure from the hydraulic pump of the fire engine. In this report, the outline of the converter is described and the experiment result of using the fire engine is shown.

View full abstract

A sarin gas incident occurred in the Tokyo subway system on March 20, 1995. This fatal gas attack caused sicknesses to many passengers. Since the disaster broke out under the ground, the main job of rescue personnel was to transport victims on stretchers to a first-aid station on the ground. So, we research and development the rescue support stretcher robot for a heavy casualty disaster in underground areas or at stations can be handled in a quick, effective manner. The stretcher robot is expected to help firelighters achieve efficient rescue operations, hi this paper, we discuss the mechanism for going up and down the stairs. And, we show the produced stretcher system with stair-climbing.

View full abstract

The rescue robots are expected to develop for decreasing human damage by natural disasters. We've proposed an Ubiquitous Security Area (US area) where is connected by Ad-Hoc network between network equipments. The US area is enable to transmit a large amount of information over a wide area and consisted of rescue robots, high performance security cameras and wireless LAN transmitters. This paper is described that the efficiency of the US area is verified experimentally in Kobe underground mall with developed systems.

View full abstract

We have been discussing a development of the gathering information system for disaster area by the wireless sensor networks. Then we propose launching deployment method of the sensor node (SN) for deploying dangerous place where a human and rescue robot cannot enter. In order to realize launching deployment, the SN has to be protected from dropping impact. Therefore, we covered SN by impact resistance structure. Prototype of SN consists of plastic ball for the impact-resistance structure and SN core for sensing, communication and information processing. And, experiment confirmed that wireless communication directionality of SN by utilizing the impact resistance structure.

View full abstract

For search and rescue mission, it is important for mobile rescue robots to navigate in unknown environments and to map them. In such situation, three dimensional information is very useful to support robotic teleoperations used in locating victims and aid rescue crews in strategizing. However, to obtain this 3-D information, the operators must wait a few seconds and halt the robot's operation using conventional scanning systems. To solve this problem, our research group proposed a continuous acquisition of three-dimensional environment information for tracked vehicles using the three-dimensional gyro-based odometry. In this research, on the basis of the above system, we developed a tele-operation method for mobile robots that allows time delay and narrow bandwidth wireless communication. In this paper, we introduce the developed tele-operation system and report experimental results to confirm a validity of the system.

View full abstract

This paper describes a local path planning and its evaluation method for tracked mobile robots. A tracked vehicle has high ability of getting over rough terrain. However, it is difficult for operator in remote place to control its moving direction and shape. We propose a new path evaluation system based on measurement of environmental shapes. In this system, the candidate paths are generated from operator inputs and terrain information. These paths are evaluated by the robot model. Finally, the combination of translational and rotational velocity is chosen. In this paper, we present a method for detecting the area where the robot cannot pass through and selecting safe paths.

View full abstract

Continuous precise procedures are strictly required in microscopic neurosurgery. Holding the instrument in the best and identical position is important, and it is achieved by a surgeon's arm itself. The aim of the project is to develop an intelligent arm holding manipulator which follows surgeon's hand automatically. We have designed and made an auto-adjusting arm holding manipulator for microsurgery. This manipulator has designed based on three concepts. 1. The manipulator should be moved freely only by the movement of the surgeon's forearm. 2. Do not use an active control to prevent the movement not intended. 3. The save operation should be done smoothly. To realize these concepts, we have designed the 5 DOF passive control manipulator with powder brakes and tension springs, and this manipulator has Fixed mode, Controlled mode , and Free mode. In near future, we will evaluate the effectiveness of using this manipulator quantitatively.

View full abstract

This paper describes the development and experimental result of master manipulator for the workspace creation in minimally invasive brain surgery. The master manipulator was development for the purpose of maximizing the doctor's facility for performing operations. Therefore, the master manipulator needs to be easy to use for the doctor. This master manipulator which was developed has three degrees of freedom to control the Workspace-creation manipulator that has eleven degrees of freedom. However, there was a problem of not allowing the slave to do complex operation in this master easily. In this research, master who was able to have the slave do complex operation was developed. This master has four degree of freedom.

View full abstract

In this paper, a 5 DOF (3 translations and 2 rotations) parallel mechanism D-8 is proposed. D-8 is an extension of DELTA-4 mechanism, which is a translational 3 DOF redundant parallel mechanism, proposed in our past study. The most advantageous benefit of DELTA-4 is a good working area distribution at the front of its structure. In D-8 structure, an end-effector bar is held by a pair of DELTA-4 mechanism which is located in face-to-face. D-8 mechanism has an advantage on its working area comparing conventional haptic devices, which have multiple translational and rotational motions with parallel mechanism. A prototype implementation of D-8 as a pen type master manipulator for surgical application is introduced in this paper.

View full abstract

In this paper, we developed a tele-operation robotic system for fingertip replantation surgery, which is called MiRS (Microsurgery Robot System). There are some difficulties on surgery process, because of tiny blood vessels and nerves of the finger. To cover the problems of surgery, the system, which is called MiRS (Microsurgery Robot System), must have two rightful prerequisites: the accuracy and comfortableness. For accuracy, which guarantees safety also, we developed a master-slave surgical system. Safety problem contains the reduction of hand-tremor on surgery, so "motion scaling" can be realized by master-slave system. Besides, motion scaling can improve the accuracy itself, either. Lastly the master-slave system with scaling motion improves comfortableness either. Most importantly, the wrist motion is added in end-effector for our system. That motion can helps make more natural and smaller movement on surgery. The process of surgery can be more safe and accurate by making small movement. We developed slave system now, and experimental results are included to verify the performance of the robot and advantages by using wrist on the end-effector.

View full abstract

In this study, authors aim at developing microsurgery tool in the next generation by rapid digital manufacturing technique. The full use of hybrid stereolithography method make it possible that the complicated, high-density, and tiny structure can be made by intermittent embedding metals or other materials into the photo-cured resin matrix on the way of the stereolithography process. In the first half of this report, the pullout strengths of original hybrid test bars were verified to have the enough strength for clinical use. It could be confirmed that the tendency of the plots of interlaminar shear stresses, for the different implantation lengths of the stainless steel tube, was equal to the theory of strength of composite materials. In the latter half, the new layout design and geometry modeling of the microsurgical tool, especially the design of a flexural joint with the fiberscope, were performed for the future utilization of the hybrid stereolithography technique.

View full abstract

This paper describes the development and experimental result of master manipulator for the workspace creation in minimally invasive brain surgery. The master manipulator was development for the purpose of maximizing the doctor's facility for performing operations. Therefore, the master manipulator needs to be easy to use for the doctor. This master manipulator which was developed has three degrees of freedom to control the Workspace-creation manipulator that has eleven degrees of freedom. When experimenting on in vivo, the opinion that the function to recognize contact with the brain was necessary was obtained. In this research, because the doctor can use it more easily, it proposes the function to recognize contact with the brain.

View full abstract

We have been researching on the remote operated robot in which it is difficult for operator to understand the situation around the robot only by the camera image for remote control. The adequate security and maneuverability cannot be secured, and danger to cause accidents increases. As the solution, we have proposed the methods to construct the environmental map to present the operator the situation around the robot. The performance trial of two range scanners was also executed and reported.

View full abstract

With the touch panel type interface, the operator cannot only look environmental information around the robot but also control the remote robot and camera by using the button on the touch-screen. In addition the operator can direct the robot to move along the same course as drawing on the environmental map on the touch-screen. Both touch screen and joystick interfaces are experimented on a developed mobile robot remote operation system and discussed on maneuverability and operating accuracy.

View full abstract

The number of the human-coexistence type robot which operates nearby people is increasing. Then a comprehensible and safety interface to operate is required by people who operate advanced robot for the first time. This paper shows a trial interface via voice to operation a mobile robot. It reports the examination to develop a voice interface, especially on mounted functions and the result of the experiments comparing with the keyboard and the step-on interface.

View full abstract

This paper discusses a mobile robot teleoperation system via an Multi-Robot Sensor Network (MRSN) that consists of Wireless Sensor Network (WSN) and multiple mobile robot. For this purpose, the teleoperation interface server that controls communication between an operator and robots via WSN was developed and a modified routing protocol based on AODV-uu was installed to MRSN. The operator can teleoperate mobile robot directly and indirectly using this system. The teleoperation system using two mobile robots and WSN composed of five small Linux PCs was developed, and teleoperation using this system was demonstrated. Communication performance was confirmed to evaluate proposed protocol.

View full abstract

This research proposes home appliance service that network robots provide. Recently, most of home appliances aren't network-reachable. Our home robot can approach a home appliance and control it. The robot notifies other robots how to control home appliances, so that others can use home appliance service via network. We make Interface Robot to connect legacy home appliances with network based on this idea and evaluate it.

View full abstract

This paper deals with a unified system of fully distributed meshed sensor network and mobile robot cooperation which serves as a sink node. A novel communication timing control employing coupled-oscillator dynamics, named Phase Diffusion Time Division method (PDTD) has been proposed so far, aiming at realization of ad-hoc collision free wireless communication network. In this paper, We extend the basic PDTD so that it can exhibit flexible topological reconfiguration according to the moving sink node(robot). Unlike conventional sensor network, no static sink node is supposed inside the network, however a mobile robot will function as a sink node and access to the mesh network from an arbitrary position. Large scale experiment was conducted and its results show that satisfactory collaboration between the mesh sensor network and the mobile robot is achieved, a≪d the proposed system outperformed CSMA-based sensor network system.

View full abstract

Human hand is one of the best manipulators existing now. When we designing Robot hands, it is available that imitates Human's one. So we made a "Dynamics simulator by the detail human model" that Musculotendon model based on real human, and looked at its behavior. In this way we try to know characteristics and construction of Human hand. Presently 34 muscles are mounted on this model. In this paper, the supinator was mounted in the detailed human mode to add the supination of the arm of hand. And its behavior was verified by comparing with the action by an anatomy book.

View full abstract

There are many lifelog researches to give services with personal daily digitized information.. To realize lifelog services, the labeling technology to enormous quantity of information is necessary. In this paper, we propose a versatile labeling technology focused on the unordinary/ordinary situation of daily life. Our approach regards the difference of data distribution collected in each action. And based on the time-oriented features of the data, we brought in smoothing process, and label-modify-process. We evaluated the accuracy of our approach with heart rate sensor data. The evaluation was conducted on three subjects. Then, we comfirmed more 13.5 % accuracy improvement with above two processes, than without them.

View full abstract

The motion analysis and the estimation of somatosensory information (including muscle tension) have been researched, and these techniques can be applied to assist the sport training and rehabilitation. For these applications, the realtime estimation and visualization of motion data and somatosensory information is necessary. The existing method is, however, computationally expensive. In this paper, we propose a new algorithm to estimate the muscle tension and pipeline computation for that to estimate and visualize muscle tension in realtime using the optical motion capture system, electromyogram (EMG), dynamic muscle model, and dynamics computation. This system overlays a rendered musculoskeletal human model on top of the images from a standard video camera attached to the display, and the user feels as if seeing his/her muscles with tension information though the cloth and skin. The proposed method realize the estimation and visualization in 16 frame/sec, which is felt smoothly.

View full abstract

In the research on the speaker array system, observation of the spacio-temporal sound pressure distribution in 3D space was difficult for the reason which requires a lot of time for measurement. In this research, in order to observe sound spot formation process in 3D, automatic measurement system was developed, and formation process of the sound spot by speaker array system was observed using the developed measurement system.

View full abstract

For providing context-oriented services for user by cooperating among robots which exist in different environments, we propose context-generation method. Based on the multiple sets of 4W-information collected by unconscious robots, an interaction results between a target user and a visible robot can be merged into the context of the target user. The proposed method is implemented into the network robot platform and a guide-service experiment in the shopping mall has carried out.

View full abstract

Assuming the guidance by robots in a shopping mall or an amusement park, we aim to establish basic technologies for the network robot platform. Using our platform, we have carried out many field experiments, "souvenir picture service", "exhibition guide service", "health support service" and "guidance in a shopping mall". Through these experiments, we have confirmed the feasibility of the platform, from the viewpoint of extendibility.

View full abstract

We propose software architecture for autonomous robots which improves efficiency of development and performance of robotic system. An autonomous robot is complicated system formed by integration of many technical elements. And, without appropriate software platform, development efficiency and system performance will go down. Therefore, we have built the software architecture which has capability for integration, extension, reuse, date sharing, distributed processing and seamless simulation with network technology. By this software architecture, we establish unified, flexible development, and reduce strain of development. In this paper, we describe about the software architecture and indicate its availability through case-studies.

View full abstract

It is well known that CORBA is a novel programming architecture and technical specifications for constructing object-oriented network systems, which is based on vendor-independent architecture and infrastructure that computer applications use to work together over networks. This study presents a new network control system for robots, in which multi clients can access simultaneously to a single server machine that implements a robot system. This CORBA-based robot control system has a potential to realize a completely-parallel processing system such as human control system. This paper firstly elaborates on characteristics and specifications of the CORBA, and indicates the necessity for the object-oriented network computing technology. Consequently, this paper shows how to make a code for programming remote-robotic control systems on the basis of the CORBA object-oriented architecture. Finally, we demonstrate that control commands that simultaneously sent from multi clients can easily refer to remote functions placed on the server without using Socket programming that is generally based on the low-level network applications.

View full abstract

In this paper we propose the simultaneous position estimation method of user and deployed sensors which are attached to objects in indoor environment. To estimate the positions of both user and sensors, the method combines user model and SLAM. Utilizing the user model, user position is calculated from not only current sensor data but also data set which is selected from log data based on the reliabilities B. Then, the positions of user and sensors are simultaneously estimated based on the SLAM. We also describe the experimental results to evaluate the proposed method.

View full abstract

The objective of our research is to develop a wireless sensor network construction system by teleoperated mobile robot. Our proposed system is that mobile robot carries the sensor nodes and deploys them for constructing wireless network in the working environment. For deployment of the sensor nodes, we introduced to estimate data transferring speed based on statistic approach. When estimated lowest limit of transferring speed is over required one, the operator can determine to place the sensor nodes and organize wireless sensor network. As the experimental result by prototype system with proposed method, it was confirmed that data transferring speed on constructed network could be maintained over required one.

View full abstract

We proposes intelligent communication modules that enable the human-robot interaction for guidance robots in public environments such as shopping malls. These modules consist of 13 modules including the module which detects the position and the behavior of people, the module which estimates the utterance period based on the facial images, speech recognition module, speech synthesis module, interaction control module, gesture synthesis module, etc.

View full abstract

Conventional RT-Components (RTC) require PC-level computational capacity and memory. To utilize RTC on embedded CPU based devices with other generic RTCs, the RTC-Lite component, which depends on a full-spec proxy component running on a PC, is necessary. For the module independency and autonomy, RT-Middleware should support embedded devices. In this paper, efficient RT-Component development and deployment process for the embedded Linux based system is proposed. Experimental results and an application for proposed embedded RTC are also shown.

View full abstract

Conventional RT-Middleware (RTM) / RT-Components (RTC) require PC-level computational capacity and memory. To utilize RTC on embedded CPU based devices with other generic RTCs, the RTC-Lite component, which depends on a full-spec proxy component running on a PC, is necessary. In this paper, RT-Middleware (OpenRTM-aist) is ported into μITRON OS for embedded systems, and its new architecture for embedded systems is proposed. Experimental results and performance evaluation are also shown.

View full abstract

This paper aim at verification of the validity of the distributed object technology in robot development, and verification of the robust person following algorithm in indoor and outdoor real environment. We propose the technique for realizing person following robustly in real environment. The proposal method is the target person detection method which combined image processing which used update HS Histogram matching, and the LRF processing using nearest neighbor cluster search. We connected a person following module to SegwayRMP and Enon. We performed verification of operation in indoor real environment, in outdoor real environment, in real environment which moves to indoor from outdoor environment.

View full abstract

In the software development for robots, accompanied by a high risk with introducing a test program into the system. Such as robot's breakdown or accident by unexpected action. In order to reduce these risks, it is important to perform verification of operation on virtual environment. In this research, we use a Simulation environment (OpenHRP) that is controlled by Software (RT component) that is made to control robot and modularized according to the function, and compare a vision simulation with a virtual environment with texture and without texture.

View full abstract

Life environment observing system using RT-Middleware is developed in order to verify the effectiveness of Open RT Platform. This paper describes development of the steerable sensor head with pyroelectric sensors and an all-around view camera, and an experiment of detecting a human using it.

View full abstract

This paper introduces "Path Planning Tool" which is under development as a part of OpenRT Platform. It provides a collision-free path planning function on the 2D plane in the 3D environment. Since its planning engine is implemented as a RTcomponent, it can be used in two ways, designing paths offline through GUI and generating path online by accessing from user's RTcomponent. The engine consists of three kinds of algorithms, mobility, planning and optimization algorithm. Several basic algorithms are built-in to make it possible for users to start planning quickly. But if those algorithms are not enough, users can add their own algorithms easily.

View full abstract

We are developing module software for realizing intelligent robotic behavior under NEDO project 2007-2011. On January 2009, in order to demonstrate the current status of the developed software, several simulation and real experiments were conducted. In this paper, the demonstration focusing wheel type mobile robot autonomous navigation is introduced.

View full abstract

In this paper, we report our activities of standardization in OMG especially the OMG Santa Clara Technical Meeting held in December, 2008. The goal of this paper is to start a technical discussion in the robotics and mechatronics community to promote RT-middleware technology (robot systems designed by reusable building blocks) that results in efficient development of complicated robot systems. As a vehicle for the rapid development and diffusion of this RT-middleware technology, we've just started up an international standardization activity on robot technologies. In order to increase interoperability, compatibility and reusability between the various robotic common components, Object Managing Group (OMG) established Robotics Domain Task Force (Robotics-DTF) in December 2005. The Robotics-DTF is looking to begin a dialog with vendors, end users, researchers, robotics organizations and other interested parties to lay the groundwork for a common platform-independent model of robotics software development.

View full abstract

The paper describes presentation methods for action indication systems. Easy-to-understand presentation is required for a person in panic in emergency situations. Presentation methods using audio speakers and projectors have been proposed for a real-time action indication system. Audio and/or visual indications are considered and an experimental system is constructed. Basic presentation methods using audio and/or visual indications are evaluated by experiments.

View full abstract