In this paper we present an AT programming organised around RoboCup: a soccer simulation system. The course participants create a number of software agents that form a team, and participate in a tournament at the end of the course. The use of a challenging and interesting task, and the incentive of having a tournament has made the course quite successful, both in term of enthusiasm of the students and of knowledge acquired. In the paper we describe thestructure of the course, discuss in what respect we think the course has met its aim, and the opinions of the students about the course.
This paper describes a new software architecture to control an autonomous mobile robot. In order to construct a multipurpose robot which is able to do various tasks in an office or a factory, it is necessary that the architecture of the robot control system should be changeable according to a given task, and the rules of the behavior should also be changeable dynamically. In order to realize such property, the authors developed a new architecture for the communication between software modules, each of which controls individual part of the robot. The architecture is based on ‘event exchange’ under the controlling of an event forwarding server, called ‘event pool’. With this architecture it is possible to re-construct the control system of the robot easily and to make the robot to adapt to its environment.
Periodic stimulation sometimes stabilizes dynamic movement of mechanical systems. In this paper, we investigate the dish spinning trick from a view point of such a periodic stabilization. The system is simply described by using a spring-dumper model of the supporting rod which has two degree of freedom. In our analysis, the movement of the dish is stabilized when the elastic rod is shaken in higher frequency than its native resonant frequency. Our theory is checked by letting a robot play the real dish spinning.
In this study, we report about mechanisms of a biped walking robot ISR-1 (Inner Skeleton type Robot 1) . We have developed a light weight robot applied the inner skeleton frames and cooperative motion of two or three actuators for one degree of freedom. Total degrees of freedom are 18 degrees. Each joint of the robot is driven by 16 actuators. Main actuator is screw cylinder type actuator that gives liner motion. This robot equips with two gyros on head and six force sensors at feet for balance control. We have confirmed walking possibility of the robot.
This paper focuses on a planning method for an iterative transportation task by cooperative mobile robots. This task requires the generation of appropriate robot paths and the formation of groups of cooperating robots. In order to realize efficient transportation, the planning architecture consisting of “Path-Generation Phase” and “Strategy-Making Phase” is proposed. The former phase generates robot paths from global environmental information and produces a graph network from the derived robot paths. In the latter phase, every robot learns a behavior strategy based on the derived graph. An asymptotic strategy-making method is used here. The global transportation strategy derived by the proposed architecture can be divided into three types, and simulation results indicate that forming the global transportation type realizes efficient transportation. Experimental results indicate that the proposed architecture is practical enough for real situation.
The advanced manipulator is designed based on systematized method for hand tool operation. It has 7 D.O.F. ultra light weight arm with 4 fingered and 14 D.O.F. hand. Tactile sensors, covered by soft rubber which can keep sensing accuracy and recognize proper sensing points, are installed on finger tips and middle frames of fingers. Two force sensor cells are also assembled in each sensor. These fingers are drived by snake tube and ropes and fluid coupling connecting to AC motor actuators. Reduction gear ratio of the finger and the arm actuators is reduced in order to made force or dexterous control easier. This paper describe how the hand and arm manipulator should be designed refining its function or using advanced technology.
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.
A video-rate stereo machine is a dedicated hardware device for range measurement. It can generate a dense range map and an associated intensity image at 30 frames per second. This paper shows a camera geometry correction module which is designed and implemented as an circuit module to improve the performance of the stereo machine. The main functions are input-image transformation and selection of candidate points of a corresponding point. The stereo machine integrated with the camera geometry correction capability not only corrects camera geometry and lens distortion, but also operates with multiple cameras positioned arbitrary, resulting in an improved measurement accuracy.
We have proposed a concept of two-fingered micro hand, and designed and built a two-fingered micro hand based on parallel mechanisms. In this paper we mainly discuss a master device to control the two-fingered micro hand and a strategy for effective operation. We observed actual human finger motions and designed a master device and a operation strategy suitable for the two-fingered micro hand. By using the developed device, we could obtain good performance of micro manipulation.
In this paper, we describe the attitude control of the biped walking robot model (a tumbler system with joints) intended for walking on an unleveled land. The sole of the foot of the biped walking robot may not always get hold of the ground on an unleveled land. Therefore, we have devised, as a walking robot model, a tumbler system whose foot has an arc sole. We have made the model, and attempted an experiment of stabilizing control by detecting the attitude using the gyroscope. The robust control system design based on the H∞ control theory, that is, the loop-shaping method is applied to the design of the stabilization compensator of the feedback control system. In the experiment, first, the attitude of the tumbler with its knee joints fixed is controlled, and then, with its knee joints of the free legs in bending and stretching movement, the robustness of the control system is investigated. As is clear from those experimental results, we can say we have succeeded in the robust stabilization control of the attitude of the walking robot model, that is, the tumbler system with joints.
Some types of parallel wire robots have been already proposed so far. Those parallel wire robot are useful for fast motion, heavy load and so on. Since each wire generates one directional force (tension), parallel wire drive system must be redundant actuating. In this paper, we analyze stability on a motion control scheme in wire length coordinates for parallel wire drive system. Taking account of such redundancy on actuation, motion stability is analyzed.by using a Lyapunov function. Basing on “Vector Closure”, it is proven that each wire length converges to the corresponding desired value and the internal force also converges to desired one. Finally, we experimentally investigate the motion accuracy on the proposed control scheme by using a parallel wire robot with seven wires.
A new approach to the modeling of plain knitted fabrics is presented for their deformation control. There exist many manipulative operations that deal with deformable soft objects such as clothes, papers, wires, and so on. Most of these operations are still done manually by skilled workers and automatic control of these operations is required. Especially, automatic handling of knitted fabrics is desired in garment industries. Automatic handling of knitted fabrics is, however, difficult to be performed by machines since the knitted fabrics can be easily deformed and extended during the handling operations. It is needed to control the deformation of knitted fabrics during the operations. A mathematical model of the fabrics will enable us to derive efficient strategy to control their deformation. Therefore, a mathematical model of the fabrics is essential for the deformation control. In this paper, we present a mathematical model of plain knitted fabrics based on their loop structure. Plain knitted fabrics consist of knitted loops and shapes of the fabric can be derived by computing the deformation of the loops involved in a fabric. It is, however, difficult and time consuming to calculate all deformed shapes of a huge number of knitted loops involved in a fabric. We thus propose a new procedure called representative loop method so that the deformation can be computed in an appropriate time. In the representative loop method, some representative loops are selected in advance in order to compute the deformed shapes of all other loops by interpolating those of the representative loops. The validity of the proposed method is then demonstrated through the comparison of the calculated shapes and the real images of deformed knitted fabrics.
Position control of an underactuated manipulator that has one passive joint is investigated. The dynamic constraint caused by the passive joint is second-order nonholonomic. Time-scaling of the active joint trajectory and bi-directional motion planning from the initial and the desired configurations provide an exact solution of the positioning trajectory. The active and passive joints can be positioned to the desired angles simultaneously by swinging the active joints only twice. Feedback control constrains the manipulator along the planned path in the configuration space. Simulation and experimental results show the validity of the proposed methods.
Newly advanced active endoscope named “Hyper Endoscope” with hyper redundant degrees of freedom for future minimally invasive surgery in the abdominal cavity is developed. The first prototype of Hyper Endoscope equipped with several “Active Universal Joints”driven by miniature cybernetic actuators is constructed and controlled successfully. In order to achieve force sensational control of hyper endoscope, a micro master slave control system with force feedback is verified by introducing “dither” technique. This system is very useful for remote surgery as well as laparoscopic surgery.