In the RT-Middleware, a data-centric communication method between RT-Components(RTCs) called "Data Port" is provided. Many types of the data-centric communication schemes, which depend on both ends RT-Component properties such as execution method, data producer's/consumer's processing throughput and period and relationship between them, can be possible. In this paper, we discuss about the data-centric communication scheme between RTCs and its new architecture is shown.

View full abstract

To apply for the RT-middleware to mechatronics, we are interested in reliability or system-size. At present, the RT-middleware needs a PC system, thus, it is important to examine the scale of RT-components. In this Paper, as one of the minimal scale, we try to provide RT-components that provide basic I/O service by PC-based direct hardware control. First, RT-components, which wrapped DA and Counter PC-interface-board, are proposed. These RT-components have data-ports for basic I/O, service-ports for ordering application service, and configurations for PC-board unique setting. As an example system, we make a simple DC-motor control system with an RT-component for very basic control.

View full abstract

We have been developing the distributed database service for the intelligent robot software platform in "Development Project for Intelligent Technologies of Next-generation Robots" of Ministry of Economy, Trade and Industry in Japan since FY2007. This database service stores and manages the robotic intelligent modules and its profiles. We consider that this database service improves the efficient of robot system development. In this paper, we introduce features, software architecture and future plans of this database service for intelligent robotic system.

View full abstract

The RT-middleware is usually operated only on PC. This restricts development of small RT devices. This research aims at the operation of RT-middleware on a low-performance MPU and the support of CAN communication interface. This makes it possible to construct the distributed robotic system using CAN network very simple and flexible. For this, we adopted CANopen by CiA (CAN in Automation) to develop practical robotic system,. In this paper, we describe development of RT-Middleware-System for Distributed-Control-System-Robot.

View full abstract

In recent years, attentions have gathered for various robots. A robot is wanted to carry out missions safely and certainly. Therefore, it is necessary to do some examinations. Therefore, in development, the module is managed in the form which includes hardware, and the motion which increases the reusability to other robots is activating. Moreover, also in mission, state grasp and system change are indispensable to failure of a machine, un-suiting to environment, etc. Generally it is necessary to absorb an obstacle by changing a function by software, and to carry out missions.

View full abstract

This paper propose the technique to build cooperation system for multiple manipulators using Robot middleware. Cooperation system is composed of manipulators,end-effectors, sensors. Each of element of a system is developed as RT-Component. We developed a resident component for realtime operation, a inteligent component for advancement of a function. This paper show a construction example of cooperation system.

View full abstract

This article proposes a systematic approach to developing reusable task skills with RT-Components. First, how to modularize a task skill is discussed to improve the reusability of the skill. Then, how to realize the modularized task skill with RT-Components is described. Finally, the methodology is applied to a nut attachment task skill.

View full abstract

To efficiently construct perception-action system for many different robots trough many trials and errors, system structure should have modularized and developmental architecture, and prototyping easiness. For realizing such software platform, RT middleware and Lisp architecture are integrated in this paper. We extend the existing implementation of the RT middleware and make it be able to construct and manage the components interactively from Lisp interpreter. For confirmation of our proposed system, several perception-action systems for different types of humanoid robots are constructed and some experiments using actual robots are done.

View full abstract

The Linux real-time preempt kernel patch, developed by Ingo Molnar, has recently drawn attention as an alternative to other dedicated real-time OSs. The patch implements real-time behavior by allowing nearly all of the kernel to be preempted, and by usage of high resolution timers. We describe our experimental environment for real-time control of a PA-10 robot arm and present experimental data for real-time capability of the real-time preempt kernel patch based controller.

View full abstract

Robots need various technologies. For that reason, it is very important to consider with reusability. In this study, we propose interfaces for an image recognition RT component and a mobile robot RT component to improve reusability. Additionally, we introduce an attendant service robot with RT components based on our interfaces. We will show that our interfaces are useful for constructing robot system.

View full abstract

In this research, we inspected a new self localization method for outdoor navigation. This method is named DFIT (Dual Floor Image Tracking). DFIT is visual tracking of floor image with two cameras. Indoor evaluation of DFIT showed its effectiveness. By introducing Increment Sign Correlation and adjustment of shutter speed to DFIT, DFIT can be applied to outdoor use with robust image data processing. As the result, it is expected the robot to provide better service. In this paper, we describe solutions to problems and result of experimentation.

View full abstract

Efficient operation of the electric power is necessary for service quality (QoS) maintenance of the robot. Therefore, we need to control the energy consumption and the remaining capacity of battery. In this research, we developed construct the capacity monitoring system for the battery. This system calculates the accurate remaining capacity considering some specific change by the output current, ambient temperature and degradation.

View full abstract

The purpose of this research is proposal and development of the system for self-preservation of the robot. This system is to be applied to the power management of dispersion control system. The observation of electric power and the control of energy-saving are mentioned in this paper.

View full abstract

This paper describes design and release plan of four softwares for mobile robot navigation. These are SSM for sensor information sharing, OGMM for multi sensor obstacle detection, POET for multi sensor localization and mapping, and SH-Spur for PWS type vehicle controll. These softwares are designed for a normal linux operating system. The sensory data processing software framewoks (SSM, OGMM, POET) are designed to modularized manner. These frame works are divided common parts of sensory data processing and depending parts of sensory data processing. These softwares abstracts unified information from multimodal external sensors.

View full abstract

In this paper, we report our activities of standardization in OMG, especially the OMG Washington DC Technical Meeting held in March, 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 small flight robot which made a flapping of wings the model is manufactured. Some kinds of mechanisms of a flapping of wings are examined, wings area is changed, and lift is measured. Furthermore, paying attention to the size and form of a lattice of wings, it analyzes using finite element method using the piece model of an examination manufactured by 3DCAD about lateral bending, longitudinal bending, and torsional.

View full abstract

This paper presents autonomous flight control of flapping-of-wings robot. The positions of the robot and destination point are measured by GPS sensor. We propose a simple control algorithm to make the robot approach to the destination point. In order to evaluate the algorithm, we show the computer simulation which takes influences of the measurement error of GPS and the change of direction of the wind into account.

View full abstract

This paper presents an innovative variable-wing mechanism based on flapping motion of birds. The main feature of the mechanism proposed in this paper is to realize the complicated flapping motion of birds such as flapping, spanning, lead-lagging, feathering by using only a single motor. We show the utility of the new variable-wing mechanism by comparring the motion generated by the new mechanism with the typical flapping motion of pigeons.

View full abstract

We have developed a battery-powered ornithopter to investigate the kinematic variation as the stroke amplitude changed. The ornithopter was designed to have 110 mm wing semi-span so that Reynolds number was the same order as that of insects (Re 〜 10^4). Free flights of the ornithopter in various stroke amplitudes were recorded with a high-speed camera at 1000 fps, and their motions were analyzed. We clarified that the body angle increased from 14 degree to 100 degree as stroke amplitude increased from 48 degree to 78 degree. We were also able to measure angular velocity and acceleration of the ornithopter with 300 [Hz] for 0.4 [s] during free flight using gyro sensor and acceleration sensor. The circuits weigh 0.71 [g] and 0.83 [g], which were light enough to attach on the ornithopter. On each circuit, sensor output was once stored in EEPROM, and retrieved later. We demonstrated our ornithopter with sensors was useful in measuring kinematic parameters in free flight.

View full abstract

The final goal of this study is to develop a flapping flying robot capable of hovering. In this paper, we focused on vibration caused by the flapping. We built a hummingbird-size robot and four types of wings that can be attached to the robot. The wing was driven by a D.C. motor and a crank mechanism. Then, experiments to investigate influences of wing shape on vibration were conducted. We measured movements of the robot by a high speed camera and found that wings that resembled hummingbird wing were better than the other, and the tail is useful for stabilizing the vibration.

View full abstract

In this study, we constructed a butterfly-style flapping robot with wing vein and investigated the flight characteristics for various design parameters such as swept-forward wing angle and position of center of mass (COM). Here, the wing length of the constructed hardware is 53mm, the total mass is 640mg, and the flapping frequency is 10 Hz. The results showed that the angle of attack was controlled by swept-forward wing angle and the relative position between COM and center of lift (COL), and that the model with swept-forward wing and rear-COM is more suitable than that with sweepback wing and mid-COM for butterfly-style flight.

View full abstract

This paper proposes a posture control of an ornithopter with a movable abdomen. By using a movable abdomen, it is possible to simplify the mechanism compared with conventional ones. We developed a small sized ornithopter with a movable abdomen, and conducted experiments to control its posture. The motion capture system was used to measure the position and posture of the ornithopter and the angle of abdomen. The experimental results showed that the proposed method was able to change the posture.

View full abstract

The purpose of the research is to develop flapping robots using piezoelectric fiber composites. By using the composites, actuating fibers and sensing fibers can be embedded into a wing as part of the structure and it makes flapping robots with compact and light structure. A prototype of flapping robot with flapping and feathering motions has been developed. The design and driving method of the robot as well as experimental results about thrust and lift are shown in this paper.

View full abstract

In order to prevent disasters or crimes as few as possible, or minimize the damage even if they occur, quick information gathering system for search and surveillance is required. An indoor blimp robot has the excellent information gathering capability and can perform three-dimensional movement. The indoor three-dimensional robot which is very close to men should be safe, quiet, and non-annoying. The blimp satisfies these properties perfectly. In this paper, a flexible switching system between manual operation and automatic operation through computers is developed. The automatic control system is constructed by tracking markers on a blimp from a pan-tilt camera on ground and estimating the position.

View full abstract

We are developing "connective balloon robots"."Connective balloon robots"can move in the room because it is as long and slender as the snake. The indoor environmental map is produced by using "connective balloon robots". Relative position of the unit is important when airship unit is operated. Airship units are connected by passive joint of two axes. Joint can measure the angle using potentiometer. Relativity position of the airship unit can be detected by using measured angle. By this text, I calculate the relative position between the airship units and simulate.

View full abstract

In this study, an airship control system was developed by MDD (Model Driven Development) method. In MDD method, models of the target system are constructed at first, and developing process is advanced using these models. The airship control system was modeled with MDD method, and "requirements model", "analysis model", and "design model" were made. Then the program of the airship autonomous flight system was coded on the basis of these models. Also the altitude/direction PID control system were embedded in the autonomous flight system. As a result, the system realizes accurate flight.

View full abstract

In this study, we present a remote control support system consists of hobby-class radio-controlled helicopter and microcomputers, various sensors. The aims of this system are simplification of the control operation and safety improvement. This paper reports some experimental results on attitude control in practical situation.

View full abstract

This paper proposes an assistant and training system for controlling an unmanned helicopter. The unmanned helicopter does not have any sensors which measure its position or posture. Stationary cameras are placed on the flight field. The helicopter is controlled by using a visual servo technique as follows. An operator steers the helicopter using a controller. The controller makes reference signals. The assistant system designs control signals such that the helicopter tracks the reference signals. The proposed system has the following four functions: automatic takeoff and landing, control channel selection, flight in a desired area, and automatic motion generation. They enable beginners to control an unmanned helicopter.

View full abstract

This paper presents cooperative control of two micro co-axial helicopters using vision-based control with occlusion handling. The helicopters do not have any sensors which measure their positions or postures. Eight stationary cameras are placed on the ground. The helicopter positions and postures are reconstructed by fusing the visual information. Experimental results demonstrate that automatic guidance control of micro-helicopters is performed.

View full abstract

This paper presents stable control of a small indoor-type helicopter only with internal sensors. The helicopter carries PSD sensors, Ultrasonic sensor and SDmotion sensor as internal sensors. We construct the nonlinear dynamic model of the small indoor-type helicopter and convert the nonlinear dynamics to a Takagi-Sugeno fuzzy model by the concept of sector nonlinearity. We design guaranteed cost controller via a fuzzy model-based control approach by utilizing linear matrix inequalities (LMIs). Simulation results show the utility of our design approach.

View full abstract

We study the control system design for autonomous flight of a portable unmanned helicopter. In order to realize the design, we use the bench test device which enables us to execute the flight experiment safely in our laboratory. The designed control systems, such as an engine control system, an altitude control system, and attitude control systems, are obtained by system identification and two degrees of freedom PID control. It is also clarified by the experiment that the designed control systems are effective for the helicopter flight.

View full abstract

Measurement of the heading which is free from magnetic declination by use of several GPS antennas is widely applied, but it is not suitable for small unmanned aerial vehicles because of its limit of the payload and body length. In this paper, we propose a method to estimate the magnetic declination using an IMU and one GPS antenna. The method consists of 1) estimation of the acceleration of the motion from GPS measurements and 2) estimation of the magnetic declination angle using the estimated accelerations and IMU measurements. We show that the estimation of the magnetic declination angle results in reduced QUEST, that is an eigenvalue problem of 2 x 2 real matrix. Experimental results of flight control of an unmanned helicopter demonstrate the effectiveness of the proposed method.

View full abstract

In this study, we developed a tail sitter type VTOL aerial robot, and a basic control system. The VTOL aircraft has advantages of both the rotorcraft and the fixed-wing aircraft. The tail sitter type VTOL aircraft doesn't need an extra actuator for the VTOL action like the tilt rotor type. On the other hand, because the attitude change is large, the control is difficult. The control of the VTOL aerial robot is roughly divided into three phases. The first is the take-off and landing, the second is the transition flight, and the third is the horizontal cruse flight. As the first step for controlling these three phases, we developed and experimented in the hovering control that is a basic maneuver during the take-off and landing.

View full abstract

In this paper, a control strategy is proposed for a vertical take-off and landing (VTOL) aerial robot with four-rotors called X4-Flyer. Since the X4-Flyer has underactuated and nonholonomic features, a kinematics based control law is first transformed into a chained form. Next, the transformed kinematic model is controlled by using Astolfi's discontinuous control. A backstepping method that is one of adaptive control methods based on Lyapunov methods combines the kinematics model and the dynamical model to construct the torque control of X4-Flyer. Finally, computer simulations are shown to demonstrate the effectiveness of our approach.

View full abstract

This paper presents development of a multiple parallel link rotor for flying robots and its efficiency improvment. A key feature of the developed rotor is to have three sets of variable wings that not only change attack angles according to wing positions but also are located so as to have 90 degrees phase difference each other. A 340g rotor with the new mechanism, which is categorized as a horizontal-axis rotor (cyclogyro rotor), is built up. First, we construct a simulation model to obtain lift and drag forces by considering the trajectories of the swinging wings, i.e., by considering the change of the attack angles. The simulation model agrees well with experimental results. Though the simulation results show that our developed rotor with a battery can hover around 7 [Hz] rotational frequency, the maximum rotation in experiments is 7 [Hz] due to low efficiency. As a result of improving its efficiency, we can achieve 7 [Hz] rotational frequency in experiments. This result shows that the developed flying robot with the new rotors is capable of generating sufficient lift force for flying.

View full abstract

This paper presents simulational and experimental studies on the design parameter optimization for the flying robot with pantograph-based variable wing mechanism. A key feature of the mechanism is not only to change attack angles but also expand and contract according to wing positions. In our previous study, experimental results show that the robot optimized with computer simulation can generate 140gf pay load. But the simulation model do not agree well with experimental results especially about the wing span. In this paper we construct the new simulation model considering wing deflection. The new simulation model agrees with experimental results and make it possible to optimize wing span. As a result of optimization using the new simulation model, we obtain the new optimal parameter. Experimental results show that the robot with new optimal parameters can generate 266gf payload.

View full abstract

In face-to-face communication, we use not only verbal medium but also non-verbal medium for communication. In particular facial expressions are very important for emotional communication because they show emotions and feelings effectively. Therefore it is thought that facial expressions are necessary for human-robot smooth communication. In this paper, "Face Robot" that has human-like appearance and can display facial expressions similar to human being is developed. In order to improve the humanity, the skin of the face robot is taken from the cast of female's face (i.e. life-mask). We then add and improve control points of the face robot according to features of her face. Moreover we analyze her facial expressions in order to mimic her real-time facial expressions with the face robot. We then confirm that mimicking of her real-time facial expressions and face features with the face robot are reproduced successfully.

View full abstract

A mascot robot system is developed for the casual communication between human beings and robots. Mascot robot system is an information retrieval and recommendation system through human speech in household environments, which has speech recognition modules, information recommendation engine that recommends information to human taking account of the situation, and five eye robots that present mentality expression as the main components. These components are added to a network, integrated by RT Middleware.

View full abstract

Human-robot interaction and communication is a quickly growing research area at the interaction of research fields such as robotics, engineering and psychology. This paper presents evaluation of emotion in interaction with machines such as robots. We provided two tasks to 62 subjects. The one is the task of getting the information through verbal communication with a robot, the other is the task of getting another information through using the web. And we requested subjects to answer the inquiry of emotions after two tasks. The first result is, the subjects experienced more strongly emotions in the case of verbal communication than in the case of using the web at some factors : "Anxious depression", "Fatigue", "Active pleasure", "Affinity" and "Consternation". The second result is, we could classified subjects into five groups and analyzed there contents with the use of answer the inquiry of emotions.

View full abstract

Recently, the existence of the robot approaches human's daily life. In such an environment, the robot not only follows the action designed in advance but also should perceive what the situation is, and do an appropriate action. Then, in this paper, it aims to achieve the perceptual system like humans for the partner robot that coexists in the same environment with people, and to acquire an adaptive action in an interaction. So approaching the interaction ability of the robot from the side of feelings and installing the perceptual system that models human's feelings in the robot., it aims at the achievement of the natural, effective interaction that humans is doing.

View full abstract

It is called the international society now, and the chance of communications in English has increased especially. However, a lot of Japanese are able not to exchange the conversation in English smoothly with the foreigner. This paper proposes the conversation model of the robot that readily becomes the other party of the practice of English conversation. Then when the robot as the English conversation tool is used, the following content is investigated: How does the person feel it when it is used? Then, what is it that is requested? And, what is requested from the scenario spoken as the tool?

View full abstract

A small humanoid which understands and executes commands in a multi-modal language through microphones and sensors in real time was developed. In usability tests, 30 subjects unfamiliar with the language were able to operate the humanoid without frustration and complete their tasks by talking to it, using gestures and touching it without a long learning stage. Multi-modal commands were more successful than spoken commands without non-verbal messages.

View full abstract

A small humanoid which can be tele-operated in a multi-modal command language through microphones and buttons in real time was developed. 27 subjects unfamiliar with the language and robot were able to operate the humanoid and complete their tasks by talking to it and pressing buttons without a long learning stage. Multi-modal commands were more effective than spoken commands without non-verbal messages.

View full abstract

We are proposing a new health care system using a simplified robot face. Patient's health care data can be stored in a personal computer and a robotic system can take many actions depending on the stored data. It can tell a patient what time to take his/her medication and warn them if they fail to take their medication at the appropriate time. The robotic system can communicate with a patient using facial expressions, a voice communication system and a personal computer monitor. In this paper, the robot face was designed using simple mechanisms in order to achieve a low cost system. The facial impressions of the simplified robot have been investigated with a questionnaire.

View full abstract

In the near future, robots will be used more and more in our daily life to support us. One problem is that human often are afraid of robots and do not accept them as supporters. To reduce the fear of human against robots and increase its acceptance, a robot that is supposed to interact with human in an appropriate way should be able to express its feelings. In this research, not only current stimuli but also "feelings" that have been memorized in past interactions by the robot are used to develop a method of robots facial expression, using a display or similar output devices.

View full abstract

Fuzzy coach-player system has been proposed as a robot instruction system using a voice interface. In the fuzzy coach-player system, a relation between human and robot is thought as that between coach and player. The voice instruction uttered by the coach that involves a degree-related adverb is guantified by using a fuzzy reasoning. The present paper is concerned with instructions for operating any postures of a forceps in three-dimensional space. Since it is not easy to judge the depth of a work space displayed in two-dimensional image, a lighting is considered to generate the shadow of a forceps. The effectiveness of the proposed method is demonstrated by experiments.

View full abstract

In this research, we are focusing on the physical customization of robots, and we tried to introduce "mutual" process into the customization. In this paper, we show a customizable communication robot which is installed the self-body-assertion (SBA) system which can accept / reject its user's customization. We carried out an experiment to verify the users' impression for the SBA system. As the results, we confirmed that the rejection by the robot increased "Interestingness", but decreased "Friendliness".

View full abstract

In this paper we dscribe that emotional synchronization is effective to a KANSEI communication system. We developed a robotic KANSEI communication system using entrainment of dynamics, and its effectiveness was examined by experiments of human-robot communication. The robotic emotion was entrained to human emotion by using a vector field of dynamics. The robotic facial expression using a communication robot "Kamin_FAl" was realized dynamically based on the robotic emotion. In the experiment of communication, the human impression was changed by the strength of synchronization of robot. Then we confirmed that this method could utilize human-robot communication to keep a comfortable state.

View full abstract

This paper describes an open source software system for robot audition called HARK (Honda Research Institute Japan Audition for Robots with Kyoto University). HARK consists of a lot of modules including multi-channel audio input, sound source localization, sound source tracking, sound source separation and recognition of separated speech for robot audition based on the data-flow oriented software programming environment, FlowDesigner. By combining these modules using a GUI environment, a user can easily build a robot audition system for various types of robots and acoustic environments. Through HARK applications to Honda ASIMO and Robovie with different microphone settings, we showed high software portability and reusability of HARK.

View full abstract

In Teleoperation, it is important to let operator know the situation of remote place correctly. We improve operability of Teleoperation with Force Feedback using Haptic Interface in this research. If there are a lot of tasks in the remote place, it is necessary to change Force Feedback method according to each task. In this paper, we propose variable force display method based on tasks in Teleoperation.

View full abstract