ISSN:
1572-8145
Schlagwort(e):
Metrology
;
robotics
;
Taguchi methods
Quelle:
Springer Online Journal Archives 1860-2000
Thema:
Maschinenbau
Notizen:
Abstract This study was designed to compare the effectiveness of using two measurement devices: the WATSMART system and the WILDCAT theodolite system for robot metrology. The former device is less expensive, and runs automatically. The theodolite system has a better resolution but is more expensive, and runs manually. The objective was to determine the agreement in their measurement accuracy, variation, and the resulting best performance condition for the robot (i.e. optimal factor–level combination) between the two kinds of measurement equipment. The idea was that even if the accuracy and variation of the equipment differed, if the practical aspects (i.e. measurement performance and the purpose of measurement) were acceptable, the less expensive equipment would provide cost and automation benefits. The experiment was conducted on a PRO-ARM RS-2200, five axis, open loop, joint-coordinate type, educational robot. A Taguchi L27 orthogonal array was selected for the experimentation. Seven controllable factors (load, speed, distance moved, orientation of the robot arm, direction of travel, height of travel, and the starting point), along with three interactions, were studied. The accuracy of the robot and the best factor–level combination (FLC) were computed for each set of observations. The mean, standard deviation and maximum values for the two pieces of equipment were also computed. Based on this, the agreement between the two equipment was deduced. Although some similarity existed, the results were not good enough to convince the authors that the two measurement systems were equally effective. Recommendations are made for the appropriate situations for using each of the measurement devices.
Materialart:
Digitale Medien
URL:
http://dx.doi.org/10.1023/A:1018508805175
Permalink
