Hirokazu Mayeda

Base parameters of manipulator dynamic models

A base parameter set as a minimum set of inertial parameters whose values can determine a dynamic model uniquely is investigated for a general parallel and perpendicular manipulator with rotational joints only. The notion of a base parameter is useful for an efficient and accurate identification of the manipulator dynamic model. A set of inertial parameters described in linear combinations of link parameters directly and completely in closed form is given and proved to constitute a base parameter set. It can be derived from this that the minimum number of inertial parameters whose values can determine the dynamic model is 7N-4 beta /sub 1/ (7N-4 beta /sub 2/-2 if the first joint axis parallel to the gravity vector), where N is the number of links in the manipulator, and beta /sub 1/ is the number of links connected by joints whose axes are always parallel to the first joint axis. After defining the base parameter, it is also shown that any base parameter set can be obtained from one base parameter set by a nonsingular linear transformation. The method adopted has the potential advantage of determining the relations between base parameters and manipulator motions. >

Experimental Study Of The Identification Methods For An Industrial Robot Manipulator

Two identification methods; ”stepby-step method” and ”simultaneous method” which takes advantage of the instrumental variable method are experimentary examined to estimate the parameters of the dynamic model of the PUMA 560. To evaluate the accuracy of the estimates we compare simulated trajectories with measured trajectory. It can be concluded that the two methods are neary same precise way to estimate the model parameters.

Strategies for pushing a 3D block along a wall

Pushing strategies to slide a cuboidal 3D block along a vertical wall and a horizontal floor without any rotation of the block are studied. Conditions for a pushing strategy is carefully formulated, and properties of the strategy and conditions on the block sizes for existence of a strategy are derived. To cope with uncertainty of coefficients of friction between the block surface and the wall and floor surfaces and mass and location of center of mass of the block, some robust pushing strategies are proposed.<<ETX>>

When is the set of base parameter values physically impossible

A necessary and sufficient condition has been shown for a set of base parameters to determine a inertial matrix to be positive definite for each configuration of a manipulator after continuous change of each joint variable of the manipulator is approximately considered as a finite set of discrete points. The condition enables us to judge if a set of estimated base-parameter values determines the inertial matrix to be always positive definite or not. A method also has been proposed to modify the estimated base-parameter values such that the set of them determines the inertial matrix to be always positive definite after it is judged not to do.<<ETX>>

Experimental Examination of the Identification Methods for an Industrial Robot Manipulator

The step-by-step method and the simultaneous method have been experimentally examined to estimate the dynamic model of PUMA 560. The merits and demerits of the both method are discussed and compared. To evaluate the accuracy of the estimated parameter values, the joint angle trajectories of a real motion and the simulated motions have been compared. From those results it can be said the step-by-step method is more accurate way to estimate the model parameter values than the simultaneous method. The estimated parameter values obtained by the both method will be accurate for the purpose of model based control of the industrial manipulator.

Parameter expressiop for modelling and inverse dynamics problems of manipulators

For improving the ability of manipulators or developing CAD systems for manipulators, it is necessary to develop a practical identification method for manipulators and an effective computational method for inverse dynamics problems. We have already developed a practical identification method for serial manipulators, and Luh et al. have proposed an effective algorithm for inverse dynamics problems, called Luhs algorithm in this paper. However, our method cannot be coupled with the inverse dynamics algorithm because the parameter expressions which suit our method are different from those which suit Luhs algorithm. Therefore, we must treat the above problems as different problems. In this paper, we introduce the concept of virtual parameters in order to treat both problems in a unified manner. First, we discuss the parameter form suited for each problem. Then we define the virtual parameters using the results of the identification problem to solve the inverse dynamics problem.