Environment- and task-driven tool for selecting industrial robots
Ruokonen, Antti Juhani (2016)
Ruokonen, Antti Juhani
2016
Automaatiotekniikan koulutusohjelma
Teknisten tieteiden tiedekunta - Faculty of Engineering Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2016-09-07
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tty-201608294468
https://urn.fi/URN:NBN:fi:tty-201608294468
Tiivistelmä
The problem of the research is to find a better solution for environment- and task-driven industrial robot selection process. Currently there are no tools or methods for the robot selection problem when considering an environment and a robot task. The goal was to find a solution for an industrial robot selection that takes the environment and the task into account and therefore make the robot selection process more simple and efficient.
This thesis solves an inverse kinematic problem within joint limits while avoiding colli-sions. Three tools were created using MATLAB to solve the industrial robot selection problem: Robot Selector for selecting industrial robots in the custom environment (mod-eled in OBJ-format) and task requirements, Robot Builder for creating robot model li-braries and modeling custom robots and Environment Builder for creating robot envi-ronment models in OBJ-format. Website was designed and created for distributing the tools and a source code of the tools. The tools were converted into EXE-format and uploaded to website (robotselection.wordpress.com). The source code was uploaded to GitHub.
A robot selection algorithm was tested empirically with a qualitative method and with a quantitative experiment. The results were good: An inverse kinematic solver succeeded in all 200 cases. The robot violated a collision distance in 1 case out of 200. The cause of the problem got fixed after the experiment. The algorithm was tested with 2 devices. Average processing time with a desktop PC was 3.88 seconds and with a laptop PC 11.5 seconds. Three test subjects tested the tools and created a robot and environment models after getting familiar with the tools. The average modeling time was about 7 minutes with the Environment Builder and about 5 minutes with the Robot Builder. The robot selection took averagely 4 minutes with the Robot Selector.
This thesis solves an inverse kinematic problem within joint limits while avoiding colli-sions. Three tools were created using MATLAB to solve the industrial robot selection problem: Robot Selector for selecting industrial robots in the custom environment (mod-eled in OBJ-format) and task requirements, Robot Builder for creating robot model li-braries and modeling custom robots and Environment Builder for creating robot envi-ronment models in OBJ-format. Website was designed and created for distributing the tools and a source code of the tools. The tools were converted into EXE-format and uploaded to website (robotselection.wordpress.com). The source code was uploaded to GitHub.
A robot selection algorithm was tested empirically with a qualitative method and with a quantitative experiment. The results were good: An inverse kinematic solver succeeded in all 200 cases. The robot violated a collision distance in 1 case out of 200. The cause of the problem got fixed after the experiment. The algorithm was tested with 2 devices. Average processing time with a desktop PC was 3.88 seconds and with a laptop PC 11.5 seconds. Three test subjects tested the tools and created a robot and environment models after getting familiar with the tools. The average modeling time was about 7 minutes with the Environment Builder and about 5 minutes with the Robot Builder. The robot selection took averagely 4 minutes with the Robot Selector.