This paper describes the GETEX (German ETS-VII EXperiments) joint robotics experiments, which has been conducted in April '99 at the first free-floating space robot on NASDA's ETS-VII satellite  . All the experiments have been performed and controlled from DLR's ground control system for space robotics applications which was connected to NASDA's ground station in Tsukuba. The telerobotic system combines sensor-based task-level teleprogramrning (as the basis for autonomy) with the features of teleoperation and shared autonomy. The hierarchical system structure is shown as well as the flexibility in programming and controlling each kind of space robotics application. This approach has led to a modular task-directed programming scheme, called Modular A & RController (MARCO), which provides a very flexible architecture to adapt the application-specific requirements to a given controlling scheme. A peg-in-hole experiment, using VR methods and the “vision & force” control scheme, by closing sensor control loops directly on-board (force) and via the ground control system (vision) is explained. During GETEX we conducted experiments with relevance to the behavior of ETS-VII in free motion mode in order to verify the existing dynamic models.
This paper proposes an online viewpoint planning method for a mobile robot to reach a goal position safely and quickly. When a robot passes through a narrow space, it moves slowly while carefully observing surrounding objects and estimating the distances to the objects precisely. On the other hand, the robot moves quickly to a next destination in a widely open space. By considering both the predicted positional uncertainty and the configuration of obstacles, the viewpoint is adaptively determined according to the the clutteredness of the nearby environment. The planner works online to cope with the actual error caused by the real motion. The robot motion is continuously performed with speed control between the planned viewpoints during visual processing. An experimental result using a mobile robot with stereo vision shows the validity of the proposed method.
This paper proposes the task based data exchange for teleoperation systems through communication network as an efficient transmission method of data between an operation device and a remote robot. On the task-based data exchange, the more important information according to the contents and conditions of the task which the robot performs is given priority to transmit, for example, by altering the contents of the transmitted data. We have built the experimental system in which the master arm in Tsukuba and the slave arm in Kawasaki are connected through N-ISDN and the standard techniques are utilized, such as TCP/IP, socket, JPEG, etc. A series of experimental task has been effectively carried out by the task based data exchange, that is, the crank operation which consists of grasp and revolution. The communication network with capacity limitation was used effectively and the high maneuverability in real-time with bilateral servo control has been realized. Then the effectiveness of the task based data exchange has been confirmed.
This paper presents a new type of human interface with force feedback attached to an arm, which we named Sensor Arm. The Sensor Arm can be utilized as an interface for interactive communication with the virtual environment and as a master manipulator of a tele-operation system. The number of degrees-of-freedom (DOF) of the Sensor Arm is seven, which is the same DOF of a human arm. Angle and torque of each joint can be measured. Moreover, in this system, force feedback can be realized at each joint of the human arm. To measure human force a new type of force sensor system is proposed in the Sensor Arm system.
A new concept of eye microsurgical system is proposed. The final goal of our research project is not only improving surgical instruments but also establishing a total surgical system applicable to today's most difficult microsurgery at the bottom of an eyeball. The new prototype of micro forceps, which has a joint to enlarge the surgical area and also has a built-in thin optical fiberscope to increase the dexterity of retinal microsurgery, was designed as the first step of a long range research plan. Since this micro active forceps is equipped with thin optical fiber inside the stem to obtain a lateral view of the retina, drastic improvement of fine operations can be achieved. And the design of active joint and end-effector were optimized theoretically.
Detailed measurement and analysis of the physical characteristics of human fingers are an important step towards the design of robot fingers to manipulate objects dexterously. This paper describes stiffness properties of the human fingertips under various conditions. Models to express the stiffness property are proposed from nonlinear curve fitting of the experimental results. Also, on the basis of the experimental results, it is pointed out that the contact area of human fingertips is proportional to displacement in the normal direction. As a result, it is observed that the stiffness varies linearly with the mean pressure which is calculated from the force and the contact area. By using the measurement results, it is revealed how the natural frequency in peg-in-hole tasks is changed depending on nonlinear stiffness and the mass of the peg. These properties are compared with those of artificial fingertips made of silicone rubber and silicone gel.
Locomotion of animals is said to be generated by CPG that is located at lower level of the nervous systerrl. We deal with an application of autonomous decentralized system like CPG to a myriapod locomotion robot. In this paper, we confirm by computer simulation that six subsystems generate gait pattern by the method of vector quantizat ion (VQ) under the conditions that each subsystem only have functions for controlling one leg and using local informations. The gate pattern is coded to the phase differences of each oscillator. Each subsystem learns a relation between the energy consumption and the phase difference using VQ. We succeeded in generating the gait pattern such that the energy consumptions for each leg are averaged.
In this paper, we discuss the active force closure (AFC) for multiple objects. We define two kinds of AFCs, i.e., the first kind of AFC where the finger can exert an arbitrary acceleration to each one of the multiple objects, and the second kind of AFC where the finger can exert an arbitrary acceleration to multiple objects without changing the relative position of each object. We show that the grasped object cannot always be manipulated arbitrarily even if the first kind of AFC is satisfied. On the other hand, since the second kind of AFC is a stronger condition than the first kind of AFC, the grasped objects are manipulated just like a single rigid body if the second kind AFC is satisfied. To show these features of AFCs for multiple objects, some examples are shown.
In this paper, we prove that the planar manipulator which consists of one actuated joints and one unactuated joints is completely controllable, namely, there exists an admissible path between any given initial state and any state. We also prove the planar manipulators with n actuated joints and one unactuated joints are complete controllable if and only if the first joint is acutated one. The proof exploits the mechanical properties of the planar systems.
A sonar-ring is one of the most popular sensor for indoor mobile robots, because it is simple and gives omnidirectional distance information directly. However, it is difficult to measure accurate bearing of reflecting points by a conventional sonar-ring sensor. And also, it is slow to get omni-directional information due to sequential driving of transducers for avoiding interference of ultrasound. In this paper, we propose a new sonar-ring sensor system for a mobile robot which can measure the accurate bearing angles of reflecting objects. The proposed system employs both simultaneous transmission/receive of all directions and accurate bearing angle measurement using differences of time-of-flight, and consequently archived fast and accurate measurement. We implemented a prototype of the proposed sonar-ring on a mobile robot, and experimental data show the effectiveness of the proposed method.
A multi-sonar-system for mobile robots under the influence of external disturbances is discussed. Many robots have on board more than ten sonars to detect obstacles on their surroundings. One of the most serious problem of such a multi-sonar-system is to prevent a crosstalk between sonars. Some papers reported a method that the transmitting signal of each sensor is modulated by unique binary pattern to identify individual sonar and to reduce the crosstalk. The authors have examined the probability of erroneous measurement of a multi-sonar-system whose transmitting signals are modulated by unique patterns, and proposed a method to synthesize a set of the pattern which gives the desired probability of erroneous measurement. In this paper, we extend the method to the multi-sonar-system under the influence of external disturbances. The effectiveness of a median filtering technique to reduce the probability of erroneous measurement is also discussed. Finally, some experimental results using a real robot are presented to verify the results.
Formerly, researchers who want to enter into the field of experimental study of walking robots, had to design and construct the walking robot by themselves, because commercially available walking robots are almost none on the market. This fact has been a greatest harrier for the popularization of the research on walking robots. Based on this present research circumstances. the authors have developed commercially available quadruped walking robot TITAN-VIII. TITAN-VIII was designed to acquire the characteristics of high performance of the movement, low price, simplicity of the treatment, and ability of functional extension so as to realize above mentioned objective. TITAN-VIII introduced some new mechanisms such as modular mechanism. wire drive system and so on. In this paper, we discuss about the design of TITAN-VIII, and show results of walking experiments.
This paper comments on the research of Mitsuda et al.  based on Gomi's experiment  . Mitsuda et all guided a manipulator using optical flows by the gradient-based method and a hierarchical data structure. Gomi solved stereo correspondence problems with the templete matching method by directing two cameras to the same point. The detectable ranges of translation and rotation in the experiments of Mitsuda et al. and Gomi are nearly equal although different algorithms are used. This paper also discusses the effect of linejitter on positioning accuracy and the constraint equation of the gradient-based method.