Usual repetitive control, when applied to path control of robots, cannot afford dynamical path speed changes, which correspond to changes in the reference period. This paper proposes a modified repetitive control that achieves feedforward control by switching the waveforms previously stored in several controllers according to speed changes, in an interpolating manner if necessary. An algorithm to implement the proposed method for robots is developed, and a deburring robot is actually constructed to change its path speed according to the cutting load, which is detected by the variation of tool revolutionary speed. Experimental results with test models show that the method can significantly reduce errors even when speed changes, and is advantageous to achieve accurate robots.
We have developed a CO2 laser radar for the robotic vision. The amplitude of a CO2 laser beam is modulated sinusoidally with an electro-optic modulator (EOM) . The distance is measured as the phase difference of the modulation between the detected return light from the target and the reference signal. The range image is obtained when the laser beam is scanned in two directions. The frequency of modulation of 5 MHz, 15 MHz, and 40 MHz are examined. The good result of the resolution of 3 cm is achieved at the modulation frequency of 40 MHz. Moreover, the range images through the simulated disaster circumstances have been obtained, which demonstrated the good transparency of the CO2 laser.
This paper describes a control scheme for a robotic manipulator system which uses visual information to position and orientate the end-effector. In the scheme the position and the orientation of the target workpiece with respect to the base frame of the robot are assumed to be unknown, but the desired relative position and orientation of the end-effector to the target workpiece are given in advance. The control system directly integrates visual data into the servoing process without subdi-viding the process into determination of the position and orientation of the workpiece and inverse kinematic calculation. An artificial neural network system is used for determining the change in joint angles required in order to achieve the desired position and orientation. The proposed system can control the robot so that it approach the desired position and orientation from arbitrary initial ones. Simulation for the robotic manipulator with six degrees of freedom is done. The validity and the effectiveness of the proposed control scheme are verified by computer simulations.
This paper describes a patient lifting arm for use on patient care robots. This device is designed to be installed on a locomotion vehicle for use in medical facilities. By changing the configuration from that of a wheel chair to a stretcher shape, when a slide plate can be inserted between the patient and the bed, it becomes possible to lift the patient. A belt-equipped slide plate construction has been employed to ensure that the patient is not subjected to any discomfort, and new functions, such as two-stage lifting action, have been added. Sensor based control provides automatic lifting of the patient while detecting the positions of the patient and the bed. In addition, for use as the sensor device, we developed a large-size two-dimensional tactile sensor using the pressure sensitive conductive rubber sheet.
A new algorithm is presented for reconstructing the 3 D road shape from camera images for the purpose of navigating autonomous land vehicles (ALVs) . The apporoximation that the road surface is locally flat enables us to determine a one-to-one correspondence between the two road boundaries, which together with our knowledge about the road shape (the “model” of roads) determines the 3 D road shape. In order to cope with inaccuracy of image data, a least-squares curve fitting technique is proposed. An example based on real image data is also shown.
A new method of generating an optimal approach velocity of a manipulator to its environment is presented in order to control the collision forces properly. First, analysis of a contact motion shows the necessity of contact control not only after the collision, as in previous studies, but also before it. Next, using a model of the force-controlled end-effector and its environment, forces generated at the contact are formulated as outputs of an autonomous system of which the initial condition is determined by the approach velocity. The optimal approach velocity is defined as the velocity which minimizes a performance index : the integration of the square deviation of contact forces from an equilibrium forces in the control of contact forces. A proportional relation between the optimal approach velocity and the contact force reference is derived analytically based on a mass-damper-spring model of the force-controlled end-effector and its environment. Results in simulation and experiment are presented to prove the effectiveness of the method.
The application of a homogeneous matrix (Denavit-Hartenberg Convention) to manipulator-equipped mobile robots can unify the movement expression of mobile robot and manipulator kinematics. This concept can be expected to enable enhanced operational efficiency for total system control. Proposed in this paper is a new guidance method for autonomous mobile robots operating on arbitrarily curved courses, using a homogeneous matrix system. We approximated an arbitrarily shaped course to circular arcs and straight lines to make easier to describe it by language as the objective course. The proposed guidance control method, based on the geometrical concept that a circular course with an infinitely long radius is equal to a straight course, is suited to these course variations. Results of actual driving experiments showed sufficiently practical performance by the proposed guidance method.