Placing of objects in unknown environments
Bernd Finkemeyer, Travis J. Kroger, FM Wahl
- Year
- 2003
- Citations
- 8
Abstract
Abstract. In practice, it is still state of the art to program industrial robots on-line. All goal positions exactly have to be taught by human operators. The alternative is to calculate positions and movements by means of interactive simulation tools. This off-line programming has the advantage, that no real robots are required, i.e. running applications need not to be stopped. Due to this, off-line programming substantially decreases development costs. On the other hand, this programming method introduces many difficulties limiting its practical usage today. The main problems are the inconsistencies of the off-line models with their corresponding real environments. This paper presents a new method to place polyhedral objects onto unknown surfaces. We introduce a special robot control architecture. It allows multi-sensor integration in an easy way. Skill primitives are used as programming interface. The fundamental concept is explained in detail. Experiments verify our object placing algorithm. The goal of off-line programming and automatic assembly programming becomes closer by the results of this paper.
Keywords
Related papers
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