Use CAD Data for Intuitive Definition of Robot Actions in a Skill-based Robot Application Development Environment

Conference: ISR 2020 - 52th International Symposium on Robotics
12/09/2020 - 12/10/2020 at online

Proceedings: ISR 2020

Pages: 6Language: englishTyp: PDF

Authors:
Dai, Fan; Li, Nuo; Clever, Debora (ABB Corporate Research, Ladenburg, Germany)

Abstract:
This paper describes our proposal of using CAD data to support the definition of robot actions in an intuitive way. Task-level programming should enable non-robotics experts to develop robot applications. However, defining the right task-level parameters requires still certain robotics knowledge and manual efforts. To reduce the amount of manual efforts, such parameters could be semi-automatically generated from robot application simulation environment, and CAD data could be analysed to derive the accurate assembly information. However, during industrial practice, we identified non-trivial limitations of CAD data analysis. In this paper, we first discuss the potential and limits of obtaining assembly information purely via geometric analysis. After that, we describe our approach of combining geometric analysis, 3D direct manipulation and robot assembly skill function to create the corresponding robot program.