101 A force control method using the relative motion between the manipulator and a workpiece

101 A force control method using the relative motion between the manipulator and a workpiece

908 Abstracts 098 Indefinite Inner product-Based Decomposition for Hybrid Control of Robot Manipulators P.R. Sinha, A.A. Goldenberg, pp 445-448 103...

132KB Sizes 0 Downloads 0 Views

Recommend Documents

No documents


098 Indefinite Inner product-Based Decomposition for Hybrid Control of Robot Manipulators P.R. Sinha, A.A. Goldenberg, pp 445-448

103 A Sfumulation Program for Analysis of any Type of Fluid Mechanical Systems "FLOWSIM" A. Romiti, T. Raparelli, pp 471-478

This paper proposes a geome~c approach to the problem of control of constrained tasks, based on the fact that the space of motion screws of robotic systems are endowed with an indefinite inner product. This leads to a unique decomposition of the naturally arising motions and forces of a robot end-effector into reciprocal screw subspaces of "freedom twists" and "constraint wrenches". A representative example is used to illustrate the generality of the decomposition to robot end-effectors. The invariance of the decomposition as compared to the noninvafiance of the prevalent hybrid control method is discussed, aided by an illustrative example.

Transients in fluid mechanical systems with any number and type of components are analyzed. The systems may be hydraulic or pneumatic (adiabatic or diabatic), with propagation, heat flow and friction taken into account. A particular form of the energy equation for gases takes care of any thermodynamic transformations. A simulation program makes the state of the system available at any instant.

099 Robust Solutions to the Regular Problem in an Active Dynamic Vibration Absorber System T. Mizuneb M. Morlya, K. Aroki, pp 449454 Two controllers for an active dynamic vibration absorber are designed to reduce the vibration to zero at a specified frequency and to non-resonant levels at the other frequencies. One of them contains a disturbance observer and realizes asymptotically a feedforward control of disturbance cancellation. It preserves the rel~ulation propm'ty independent of small perturbations in the ptamary system, but is sensitive to those in the absorber system. The other controller incorporates a disturbance model into its feedback loop. It preserves the regulation property in the presence of small perturbations in both the primary and abs~ber system parameters.

100 A Stndy on Control of Flexible Hydraulic Actuators for Extreme Enviremmmt Manipulators S. Yekota, T. Komatsu, pp 4~-460 The actuator proposed and developed here for hydraulic control in ofSo~ is the flexible hydraulic actuator (FFIA). The main features FHA are that it needs no lubrication and has no leakage. The precise mathematical model of the FHA in static is derived and compared with experiments. The dynamics of a single-arm model of a manipulator composed of two FI-IAs are investigated experimentally. An internal pressure control method with flow rate compensation is proposed to improve the stability and the responsiveness of the system. A disturbance estimation-observer is employed to eliminate the effects of system parameter changes and distmbanees.

101 A Force Control Method Using the Relative Motion Between the Mantlmlator and a Workplece K. WatmmBe, T. Fekud& S.G. Tzafestas, pp 461464 Control methods based on the relative motion between the manipulator and the workpiece are described for controlling the force of a one-dimensional manipulator, in which it is assumed that there are no collisions between the manipulator and the work:piece. The computed force law used is similar to the ted torque law in the trajectory tracking problem of a lator. Two cases are considered, depending on whether the position and velocity of the workpiece (or end-effector) are available or not to calculate the computed force control. The effectiveness of the proposed control methods is illustrated by some computer simulations.

102 A Geometric Approach to Control System Design of SelfSeining Magnetic Bearings T. ~ , H. Bieuler, pp 465.470 Disturbance cancellation for a special type of active magnetic bearing, the salf-sensing magnetic bearing, is investigated. The method appfied is the geometric approach, This problem is of practical as well as theoretical interest. Two types of disturbance are investigated, static load and harmonic disturbance (rotor imbalance). It is shown that the combined plant-disturbance model is observable for ainusoidal disturbance and not observable for static disturbance. As a result virtual zero power operation is achieved under static load, while periodic disturbances can be cohere,usated. Experimental results support the theoretical predictions.

104 Precision Siidin~.Mode Position Control in the Presence of Priction H £ . Chahyadl, D.C. Rye, pp 479484 A scheme for precise, high-speed positioning of a table driven by a DC motor and a fiat metal belt is described. The system dynamics are nonlinear due to the presence of friction, and because the effective belt stiffness changes as the table moves. In addition to requiring fast, high-accuracy positioning, control robustness is important. Conventional linear time-invariant control, such as PID control, is sensitive to nonlinearities. Simulation studies show a sliding-mode controller to be well matched to this nonlinear system, in terms of insensitivity to parameter variation. 105 Dual Mode PI Control for the Robot with Redundant Degree of Freedom Y. Mori, S. Nyudo, pp 485-488 This paper describes a dual-mode PI control system design for a robot used for starching work. From the features of this work, the robot has redundant degree of freedom. In this paper, the full range of movement of the robot is split into a gross motion part and a fine motion part, so as to achieve a good tracking performance. The gross motion part can move widely but slowly and roughly; in contrast, the fine motion part can move sharply and finely but its movement range is narrow. The validity of the dual-mode Pl control architecture is shown. 106 A Discrete.Time Leamnlng Control Scheme for Comtroined Manipulators Danwei Wang, Y.C. Sob, C.C. Cheah, pp 489.492 A discrete-time learning controller is presented for precise path tracking of both motion and force for constrained robots. The problem of the robustness of the learning control system to external disturbances and errors in the initial condition is studied in detail. 107 A Mobile Platform with a Robotic Arm Working In a Seml-Structurod Environment: Two Basle Contflbutious N. Clavel, P. Thompson, F. Sevila, IL Zapata, pp 493-500 The robots for rural and outdoor activities discussed in this paper will navigate in semi-structured environments and execute intelligent tasks with the robotic actuators they canT. Most actions will have to be done while avoiding stopping the mobile. Two problems generated by such a concept of the machine are addressed in this paper: how to navigate in a partially known and evolving environment, and how to combine robotic arm motions with those of its supporting mol~le platform. Tests are conducted on the specially designed and built agricultural mobile robot O']'OMAT.

108 Dextrous Manipulation of Rigid Objects Using Learnin£ Hybrid Control Techniques M. Alcardi, G. Cannata, G. Casaiino, pp 501-506

Dextrous manipulation of rigid objects involves two main objectives: to apply to the grasped object proper force/torque wrenches in order to perform an assigned motion, and to plan suitable contact forces that satisfy both the motion specification and the physical constraint characteristics of a given grasp. A formafism suitable for representing the space of the contact forces, allowing a grasped object to perform an assigned motion, is presented, for use whenever robust grasping planning problems are dealt with. The paper also shows that manipulation dexterity in grasping actions can be attained in repetitions of the task, by extending the Iterative Learning Hybrid Control Theory to robotic manipulation.