Search

This paper presents aims at mobility improvement of flexible underwater robots. For this purpose, a novel propulsion method using planar structural vibration pattern is proposed, and tested on two kinds of prototypes. The result of experiments showed the possibility of the movements for multiple directions: forward, backward, turn, rotation, drift, and their combination. These movements are achieved by only one structure with two actuators. The results also indicated the possibility of driving using eigenmodes since movements were concentrated on low driving frequency area. To investigate the relation between movement and structural vibration pattern, we established a simulation model., Comment: 8 pages, 21 figures in 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

The evolution process for creatures is very, very long, and contains many useful secrets and rationality mostly hidden in their structure, motion and configuration. Biomimetic mechatronic design is a useful approach for future mechatronic innovation, which can significantly enhance the performance of mechatronic systems. As an important issue for biomimetic mechatronic design, it is necessary to make a robot as soft as a natural creature to achieve more efficient, high-performance and creature-like motions. Compared to a conventional rigid robot, the design and control of a soft robot is difficult because the coupling between the flexible structure and surrounding environment should be considered, which is very difficult to resolve due to the large deformations and complicated and coupled dynamics. This is the main reason why design methods for soft robots has not been established, despite the many trial developments of soft robots that have been undertaken to date. The challenges for the design of biomimetic soft underwater robots based on numerical simulation considering the coupling between the flexible structure and surrounding fluid as well as control technologies are described in this chapter.

This paper proposes a disturbance rejection controller using real-time ZMP (Zero Moment Point) regulation which could maintain the stable walk of a humanoid robot when the robot is subjected to external disturbances. The controller detects the disturbance by an accelerometer, and computes the ZMP increment to overcome disturbance to ensure stability, and then calculate the increment of trajectory of center of mass according to ZMP increment using preview control. The effectiveness of our proposed method was confirmed by simulation in ADAMS and walking experiment on an actual humanoid robot, BHR-5.

Basic concept about how to realize hyper dynamic manipulation by a robot has been given by the authors in previous works. Based on the analysis results of human hyper dynamic manipulation, two key issues concerning the realization of hyper dynamic manipulation by an anthropomorphic mechanism, namely, utilization of dynamically coupled driving and structural joint stops, have been emphasized. Therefore, unlike that for conventional manipulators, our approach is to design and control a hyper dynamic manipulator based on the utilization of dynamically coupled driving and structural joint stops. Motion generation based on this approach is critical to the utilization of dynamically coupled driving and joint stops, and presents great difficulties. In this paper, a new method for motion generation is proposed, and its superior numerical stability achieved with high accuracy is shown by simulation results. Many meaningful and interesting results for the hyper dynamic manipulation for a two-joint manipulator are obtained in simulation. Moreover, the hyper dynamic manipulation is realized successfully by a prototype of a golf swing robot, designed based on the concept presented in this paper.

Recently, data traffic for satellite communications is increasing rapidly, due to various new purposes besides the conventional voice phone and facsimile. For the reason, it is necessary to realize high frequency communications, which is to be assured by high accuracy of tracking control of an antenna. To meet the needs, this paper aims at the accuracy improvement of conventional ship mounted tracking antennas by introducing H-infinity control. In this paper, the basic characteristics of tracking antenna are investigated by experiments and the model of the antenna for controller design is established. Based on the model, an H-infinity controller is designed and implemented to the antenna. Experimental results show that the tracking accuracy can be improved to /spl plusmn/0.3 [deg], which is five times higher than that by the conventional controller.

To achieve a human like grasping by the multi-fingered robot hand, grasping force should be controlled without information of the grasping object such as the weight and the friction coefficient. In this study, we propose a method for detecting the slip of grasping object by force output of the Center of Pressure (CoP) tactile sensor. CoP sensor can measure center position of distributed load and total load which is applied on the surface of the sensor within 1 [ms] . This sensor is arranged on finger of the robot hand, and the effectiveness as slip detecting sensor is confirmed by experiment of slip detection on grasping. Finally, we propose a method for controlling grasping force resist the tangential force added to the grasping object by feedback control system of the CoP sensor force output.

The realization of human-like dynamic motions by bipedal robots is still a challenge. One reason for this is that joints of bipedal robots are actuated with low backdrivability. In this paper, to improve backdrivability of the joints in a bipedal robot, a new mechanism of knee joint with continuously variable reduction ratio adaptive to load is proposed and a control method is also proposed according to the characteristic of the mechanism. A prototype of the bipedal robot using the designed knee joint with continuously variable reduction ratio adaptive to load has been developed. The effectiveness of the mechanism is shown by experimental results using the prototype.

The motion control skill in human hyper dynamic motions, especially in ultra high speed motions, is being investigated by the authors toward a realization of hyper dynamic manipulation by robots. As a case study, a new golf swing robot, consisting of an actuated joint and a passive joint with a mechanical stopper, has been developed by us. To improve the accuracy of the motion, learning control using ANN is introduced to shadow swing motion and the swing motion of hitting a ball, and its effectiveness has been shown by experiments. Since the model used in the method depends on the swing conditions, it is difficult to adapt the same model to all conditions. To improve the versatility of the learning control method, a recurrent ANN as a model with versatility is adopted to derive the model of direct dynamics of such strongly coupled, non-linear, non-holonomic and hybrid system. And a control method based on learning and motion generation by using the model is proposed and implemented to the golf swing robot prototype. In this paper, the method for adopting the recurrent ANN and implementing the learning control is described, and experimental results are shown.

A new golf swing robot to simulate human motion has been developed by authors. This paper deals with motion control for the robot hitting a ball. In the case of golf swing, it is difficult to control the impact process itself, because the time of impact between the club head and the ball is extremely short. Therefore, it becomes important to control the robot after the impact, to avoid breaking a club, etc. In this paper, a method for controlling the robot adaptable to various impact conditions, is proposed. The motion state of the robot is estimated first by an extended Kalman filter, and a new reference trajectory is generated on-line according to the estimated state. The method has been implemented to the robot, and its effectiveness is shown by experimental results.

A new kind of the rotary encoder based on the magneto-optical (MO) storage is proposed. Using the MO methods, a small high-precision high-response rotary encoder can be realized. High precision is achieved by a two-stage process in which the code disk is firstly written using the direct transfer recording, and then rewritten using the corrections based on the measured errors. The systems for writing to, reading from, and evaluating the MO encoder are developed, with the optimal MO encoder reading and writing parameters determined experimentally. A trial MO encoder with a resolution of 9000 pulses/revolutions is created directly from a reference encoder, delivering an accuracy of approximately 2.8 in. The maximum resolution and response frequencies achieved are 18000 p/r and 400 kHz. After the corrected transfer recording, accuracies of 1.3 in at a resolution of 2250 p/r and 1.7 in at 9000 p/r are achieved.

For Above Deck Equipment (ADE) on a ship, it is necessary to perform experimental modal analysis (EMA) by base excitation. Existing EMA tools are designed to perform curve-fitting on the frequency response function of displacement or acceleration at measuring point to an excitation force. However, in case of experiment by base excitation, transmissibility of acceleration, that is, the ratio of accelerations between measuring point and base is to be used for modal analysis. A method, which makes modal analysis for base excitation possible by performing curve-fitting on the transmissibility of acceleration using existing EMA tools, is proposed in this paper. First, the relation of transmissibility of acceleration using modal parameters in case of base excitation is derived. Based on the relation, a new method for modal analysis is given and its feasibility for one D. O. F. system is shown by numerical calculation. Finally, the method is applied to the modal analysis of an ADE, and its effectiveness is shown by comparing experimental results with simulation results using the modal parameters obtained by the method.

A new golf swing robot to simulate human skill of motion control with one actuated joint and one passive joint has been developed by authors. Because of the special mechanism with non-holonomic constraint of the robot and the motion of high speed, it is difficult to realize an accurate swing motion by usual feedback control. To solve the problem, this paper describes a control method using mainly feedforward control based on an approximate inverse dynamics model (IDM) close to the real system. The approximate IDM is composed of a simple IDM and a compensation IDM represented by a neural network and is to be improved by learning in doing. By this method, most part of control input is shifted from feedback control to feedforward control and an accurate swing can be realized. The method is implemented to the prototype of golf swing robot, and experimental results of shadow swing before and after learning control show the effectiveness of the proposed learning control.

As one of home services by a mobile manipulator system, we are aiming at the realization of the assist for elderly people. The purpose of this work is to develop a robot system that provides mental support for elderly people as one part of the home services. We have developed a rock-paper-scissors-Atchimuitehoi gaming system as entertainment robot system by using Kinect v2 as human robot interface. In this paper, the configuration of the developed system is described. And the basic algorithms for recognition of human action as well as voice are proposed and implemented. Finally the feasibility of the system is shown by the experimental results using the developed prototype.

In our previous work, robust control structure based on disturbance observer for a fast and precise fine motion mechanism has been proposed. The functionality of the controller was verified using one axis of precise X-Y table combined with-one axis fine motion mechanism. As a further step in this paper, the approach is extended to two-axes fine motion mechanism with a SCARA robot as a real system to realize the final purpose, that is, the development of high-precision mounting robot. First, the paper introduces a high performance robot system based on a SCARA robot coupled with a fine motion mechanism which was developed in our laboratory. Second, the paper presents a control scheme that enables correction of possible offsets of the end-effector relative to the target and compensation of the oscillations at the end-point. Finally, experimental results show significant improvements in end-point accuracy and settling time achieved by the novel configuration of the coarse/fine robotic system. As a result, a repeatability (3σ) less than 2 μm and the CP positioning error (3σ) less than 4.0 μm were achieved by the coarse/fine robotic system.

The characteristics of angle transmission mechanisms exert a great influence on the servo performance in the robotic or mechatronic mechanism. Especially, the backlash of angle transmission mechanism is preferable the small amount. Recently, some new types of gear reducers with non-backlash have been developed for robots. However, the measurement and evaluation method of the backlash of gear trains has not been considered except old methods which can statically measure at only several meshing points of gears. This paper proposes an overall performance testing method of angle transmission mechanisms for the mechatronic systems. This method can measure the angle transmission error both clockwise and counterclockwise. In addition the backlash can be continuously measured in all meshing positions automatically. This system has been applied to the testing process in the production line of gear reducers for robots, and it has been effective for reducing the backlash of the gear trains.

This paper introduces rapid obstacle sensing using mobile robot sonar (ROSUM) a new method of applying simultaneous ultrasonic firing in environmental scanning. ROSUM focuses on treating undesirable crosstalk data with a heuristic pattern classification technique. In addition, least square curve fitting is applied to estimate bearing angles to the echoes’ reflecting points which partially eliminates the problem of sonar’s poor directionality. ROSUM enables full panorama scanning to be performed more rapidly then error eliminating rapid ultrasonic firing (EERUF), currently the most rapid method available. This makes ROSUM suitable for fast mobile robots navigating in dynamic environments since the system can detect obstacles which suddenly appear in the robot’s path, and which may not be detected by the low-frequency scans in time for the robot to safely avoid them. Criteria for sonar ring design enabling multi-wave superposition to attain a quasi-homogenous sonar beam intensity pattern are also outlined. ROSUM has been implemented and tested in a specular office-like environment using a mobile robot. As a result, obstacle detection was performed not only faster, but also more accurately in comparison to conventional methods.

This paper presents a controller structure for a fast and precise fine positioner which can be attached to the end-effector of a classical industrial robot for assembly tasks. The overall control system consists of three elements: a position loop feedback controller, a feedforward controller, and a disturbance observer for the position loop. A useful method for compensating for the asymmetry and non-linearities of the piezoelectric based system is used as the first step in the design process. The robust feedback controller based on the disturbance observer compensates for external disturbances mainly from the coarse positioner. Precise tracking is achieved by the feedforward controller. Experimental results are presented to demonstrate performance improvement obtained by each element in the proposed robust structure.

Automatic garbage collection site for removing garbage from AMR and putting the garbage into garbage depot in good order has been developed. This system is composed of five subsystems: (1) A servo-controlled orthogonal type XYZ-crane; (2) A garbage bag gripper which has two rollers at the tip of its finger; (3) A sensor system to detect distribution of garbage bags on floor and on AMR; The irregularities of the garbage-bag can be detected as distance image by scanning ultrasonic range sensor; (4) A communication device using IR for communicating with AMR; (5) Computer system for control whole systems. Moreover an algorithm for determining suitable position to taking and putting garbage bag from distance image has been developed. The experiments using all of the above-mentioned subsystems have shown practicality of this system.

In this thesis, a new approach to Self-Positioning-System for an Autonomous Mobile Robot (AMR) is introduced. The system is composed of a set of small landmarks scattered inside of AMR working area and a reading system mounted on AMR. Each landmark is represented by a unique pattern made of polarizing film that is placed in a known fixed position. The panoramic scanning system is composed of a laser beam emitter and a detector of the reflected polarized light beam. When AMR approaches landmark's neighborhood, AMR can determine it's own position (x, y) and posture (orientation θ) by reading the landmark. By experiments, the system is able to determine AMR position with standard deviation of 10 [mm] or less and AMR posture with standard deviation of 1.0 [degree] of less when reading system inside of a circle (R=500 [mm] ) centered 800 [mm] away of the landmark.

Developing fish robots by biomimetic approach is a useful way to achieve more efficient, high-performance and fish-like motions. And it is very important to make the fish robots as soft as fishes. For this purpose, we are developing soft fish robots using flexible piezoelectric fiber composite (PFC) as actuators together with soft body structure. Comparing to the conventional rigid fish robot, the design and control of soft fish robots are difficult, because the coupling between the flexible continuum and fluid should be considered, which is very difficult to be solved due to large deformation and complicated dynamics. This is main reason why the design method for soft fish robots has not been established while many trial developments of soft fish robots have been done up to now. It motivates us to challenge the design and control problem of soft fish robots based on the analytical simulation with considering the coupling between flexible structure and fluid. In this paper, fundamental analysis for the design and control of soft fish robots using PFC is described. First, modeling of driving from input voltage of PFC to the displacement output of robot in the air is established and improved by comparing experimental results with simulation results. Next the method for structural design of fish robot is performed by modal analysis in the fluid. Finally, a new design of fish robot by the method is shown.

This paper proposes a design of common platform for small humanoid robots. There are many kinds of small humanoid robots for hobby or research. These humanoid robots are operated in different platforms developed by different developers. For this reason, the current development method is very inefficient. To solve the problem, it is necessary to develop common platform for all small humanoid robots. For this purpose, we are making a common platform for small humanoid robots on RT-Middleware which is used to build intelligent robotic systems. In this paper, first we investigated the requirements of the system to control motion-based humanoid robots. Then we designed a common platform and applied this platform to three different types of humanoid robots. As a result of this, the usefulness of this common platform was demonstrated.

Recently, many challenges are being made toward the realization of flight motions like insects, birds and so on. The purpose of this work is to develop flapping robots using a new type of piezoelectric material, that is, piezoelectric fiber composites. By using the composites, actuating as well as energy-harvesting fibers and sensing fibers can be embedded in the structure and it makes flapping robots with compact and light structure. This paper proposes a new flapping mechanism for the flapping robots using piezoelectric fiber composites to improve the performance of the robot developed in our previous work, which is inspired from insects with indirect flight muscle. A prototype of flapping robot using the new flapping mechanism has been made and experimental results show the improvement of the performance.

A teaching robot has been developed for the purpose of sophisticated teaching operations, such as to input technical skill of artist or artisan when machining industrial art objects. To realize easy operation, the teaching robot is constructed with a seven D.O.F. articulated mechanism and a two D.O.F. working table, in total has nine D.O.F. To improve teaching accuracy, error analysis and calibration for all geometric parameters are performed. As the results of experiments, an accuracy of 0.75mm(±3σ) by the teaching robot has been obtained. In this paper, error analysis of the teaching robot and geometric error model of it are discussed. Calibration algorithm and accuracy estimation are described also. Finally, an example of teaching and playback is shown.

A robot machining system for industrial art objects has been developed. The objective of the system is both to make good use of technical skill of artist or artisan, and to realize the automation of machining of industrial art objects. The developed system is composed of three units: (i) input unit to record the machining data of artist or artisan directly, or to measure the model of an industrial art object; (ii) data processing unit to generate and modify model from the input data, and to transform model data into machining data; (iii) machining unit to machine industrial art objects according to obtained machining data. In this paper, the general philosophy about the system is given, and the construction of developed system, design policy and functions of each unit are described. An example of machining experiments is given to show the practical availability of the system.

This paper proposes a new motion planning method for a hyper-dynamic robot which aims at realizing the motion control skills exhibited by professional golfers by a dexterous robot. The robot has similar distribution of actuators' capability to that of human, which is compact but requires dynamically-coupled driving for high speed motion. For such a robot with strong coupling and nonlinearity, to realize a motion including multi-boundary conditions, motion planning is very difficult. As a new method for the motion planning, the motion planning method using reinforcement learning is proposed in this paper. Simulation results show the effectiveness of the proposed motion planning method.

The “free-pallet” index conveyor system has been developed by the authors. This paper deals with one-axis toothed belt conveyor used as positioning mechanism in the system. A prototype of one-axis conveyor system has been developed to investigate positioning capability. To construct a flexible system which can easily change stroke, duration, velocity, and acceleration for different purposes, the conveyor is driven by a DC servo motor and controlled by a personal computer. Motion control using cam curves is applied to realize high accuracy and high speed conveying. Positioning accuracy, velocity characteristics, and residual vibration of the conveyor are also tested by experiments, and the results show that the developed conveyor system is effective as the positioning mechanism in the free-pallet index conveyor system.

Slip-detecting tactile sensors are essential for achieving human-like gripping motion with a robot hand. In previous research, we developed a flexible, thin and lightweight slip sensor that exploits a resistance change of the pressure conductive rubber. This sensor distinguishes between both contact and initial slip by using a complicated resistance change just before slippage of object is generated. In this study, we evaluate the stability of the output in the case of differing initial normal forces and changing surface properties of an object. Finally, we describe an experiment on adjustment of the gripping force by a multi-fingered robot hand.

Recently, flapping robots are noticed by many researchers, and many challenges are being made toward the realization of flight motions like insects, birds and so on. The purpose of this work is to develop flapping robots using a new type of piezoelectric material, that is, piezoelectric fiber composites. By using the composites, actuating as well as energy-harvesting fibers and sensing fibers can be embedded in the structure and it makes flapping robots with compact and light structure. This paper proposes a new driving mechanism for the flapping robots using piezoelectric fiber composites. The driving mechanism is inspired from insects with indirect flight muscle. A prototype of flapping robot using the new driving mechanism has been designed and experimental results show the feasibility of the proposed mechanism.

As one of the home services by a mobile manipulator system, we are aiming at the realization of the motion assist for elderly people. The purpose of the work is to realize a safe assist by real time feedback control based on the information from a high speed tactile sensor for detecting the force and its center between the assisted people and robot arm. This paper introduces the design of the tactile sensor as well as initial experimental results to show the feasibility of the proposed system and the control methods about basic movement necessary for stand-up movement support according to the tactile information.

A mobile manipulator system as a versatile platform for home service has been developed by authors. The mobile system has two features. One is the efficient and easy recognition of environment and objects about geometrical, physical and additional information by a sensor fusion system consisting of RFID, camera and laser range sensor. The other is such geometrical, physical and additional information-based mobile manipulation for various objects. In this paper, mobile manipulation of various chairs for home service is described. The methods for recognition and mobile manipulation are proposed and experimental results are given to show the feasibility of the proposed methods.

This paper reports our recent results on the motion control of cooperative transportation by two mobile manipulators following a reference trajectory on the slope. We propose a cooperative control algorithm which is divided into two steps. In the first step, the mobile manipulators regulate their postures to keep themselves away from falling using their redundant Degree-Of-Freedom. If the robots still cannot keep stable, then in the second step, both robots adjust their forces acting on the transported object to avoid falling. Simulation results are given to validate the proposed method.

Recently, there are many needs for small tracking camera systems to be mounted on mobile platforms, such as bikes, cars, etc., to tracking some targets stably in a perturbation environment. To meet the needs, a small tracking camera system has been developed by authors. The purpose of the tracking camera system is both to stabilize the camera and to track the target simultaneously. The system consists of three axes driven by DC motors with encoders, to form velocity and position control loop. Outside the loop, another position control loop is composed based on the integration of both rate gyro and inclinometer by a designed filter, to stabilize the camera. And besides the two loops, another position compensation loop is formed by introducing the camera position information, to track the target. This paper describes the system configuration, the design of controller as well as sensor systems, and the experimental results on the tracking performance.

This paper describes a mobile manipulator system as a versatile platform for home service. Two key technologies are addressed in the paper. One is the efficient and easy recognition of environment about geometrical, physical and additional information by a sensor fusion system consisting of RFID, stereo camera and laser range sensor. The other is such geometrical, physical and additional information-based manipulation including mobile manipulation for the heavy objects cannot be handled up by the mobile manipulator. The system configuration is described and basic experiments show its effectiveness.