transformed with new recovery configuration when a fault Failure of any component of a robotic system during occur. operation is a mauer of concern. This work investigates In this work, identification of response phenomena internal shock phenomena due to the failure of joint due to the failure of joint actuators, and then definition of actuation, and a recovery algorithm for both serial and the recovery process were studied for both serial and parallel mechanisms under such circumstances. A control parallel mechanisms. The internal shock problem was algorithm is studied that consists of a model reference simulated by assuming that an actuator failed during algorithm and computed torque method in the feedforward manipulation. Simulations revealed that several process, and a simplePR) controller in thefeedback disturbances occurupon failure. The first oneis causedby process. Simulation results illustrate the effectiveness of the shock of failure, which may be of a large magnitude. this recovery algorithm which attempts to reduce the The second one comes from implementing an emergency internal shock when failure occurs, and accomplish the function at the start of recovery such as the ,braking action tracking of the given end-effector trajectory. The outlined for the serial robot or torque redistribution for the parallel recovery algorithms, which include two stages of robot robot. The third one results from the increase of model control, path planning and path tracking, are expected to (parametric) errors. This disturbance is due to the be applied not only to a case where some joint is fully configuration change following failure. The foucdi one is failed, but also to cases where some joints experience caused by a sudden set point change since the recovery partial failure, process immediately attempts to drive the robot back to a new desired joint-space trajectory. Initially, a model.based algorithm using tlmcomputed

To design and precisely control a manipulator requires a representative dynamic model of the system. This paper presents the derivation of a rigid-link model for the serial manipulator, which reduces all of the arm’s dynamic properties to their effective values at the generalized inputs. The component terms of the model are readily calculated from the dynamic influence coefficients, which are based only on the geometry of the system. All necessary influence coefficients for serial manipulators are given in a particularly simple form. The model formulation keeps the system parameters and the input dynamics explicit in the controlling equations of motion, such that analysis and dynamic response results can be obtained in the most direct manner. Dynamic analysis results for an industrial manipulator are presented.

This work suggests a new (novel) approach for the control of taskspace stiffness characteristics in systems consisting of a superabundance of kinematically dependent inputs. When there are more input actuations than operational degrees off", intemal preloads can be generated that produce effective restoring farces in the face of displacement or disturbances imposed on the system. Examples of this excessive actuation can be found in cutain modes of structurally ovenxnstrained parallel manipulators and in antagonistically structured serial manipulators. The input loads are synthesized off-line (prior to operation) and entered as a feedforward component so that the desired effective loads on objects are obtained and simultaneously significant disturbances at the task level can be largely rejected in an open-loop fashion. This reduces the burden and shortcomings of standard feedback schemes. Moreover, a layered feedback scheme is employed to compensate for small perturbations and unmodeled dynamics. The open-loop task-based stiffness control scheme as applied to structurally parallel mechanisdrobotic lhkage systems is investigated. The scheme's applicability as a programmable active RCC devices is also discussed.

No abstract

To establish the complete dynamic model of a manipulator arm requires precise knowledge of the system mass and stiffness properties. This paper presents an experimental procedure based on the modal analysis concept to identify the dynamic parameters of industrial manipulator arms. The mass and stiffness content of an industrial robot (Cincinnati Milacron T3-776) are extracted from the modal behavior of the arm utilizing a general mathematical model for serial manipulators. The validity of the identified parameters is verified by comparison of the measured vibration behavior of the arm with the behavior of the proposed model. Finally, some suggestions are presented to improve the performance and operating speed of the T3-776 robot by increasing the structural integrity of the system.