Manufacturing Task Description for Robotic Welding and Automatic Feature Recognition on Product CAD Models
Alexander Kuss, Thomas Dietz, Konstantin Ksensow, Alexander Verl
- 发表年份
- 2017
- 引用次数
- 27
摘要
In this paper, a novel approach for an ontological representation of continuous-path manufacturing tasks is presented on the example of robotic fillet welding. The representation specifies manufacturing information and product related geometric constraints relevant for automatic program planning in robotic arc welding. The geometric constraints are transformed into simple continuous curve primitives such as lines, circular arcs and B-spline curves supporting the easy integration in robot program planning systems decoupled from a complex CAD environment. It also represents information on welding gaps along the joint geometry. Moreover, a method for automatic feature recognition based on rule- based pattern recognition is presented to automatically retrieve manufacturing information from the CAD model of the product and generate the respective manufacturing task specification. The method is evaluated in several test scenarios for welding product assemblies indicating the usability and effectiveness for automatic program planning in robotic welding.
关键词
相关论文
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Fractional Differential Equations
Igor Podlubný
2025
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991