Programming industrial robots from few demonstrations.
| Type of publication: | Book chapter |
| Citation: | Calinon_IET_2023 |
| Booktitle: | Human-Robot Collaboration: Unlocking the potential for industrial applications |
| Edition: | Doulgeri, Z. and Dimeas, F. |
| Year: | 2023 |
| Pages: | 9-37 |
| Publisher: | Institution of Engineering and Technology (IET) |
| Abstract: | A wide range of industrial applications can benefit from robots acquiring manipulation skills by interaction with humans. This chapter discusses the challenges that such learning process encompasses, including the development of intuitive interfaces to acquire meaningful demonstrations, the development of representations for manipulation skills exploiting the structure and geometry of the acquired data in an efficient way, and the development of optimization and optimal control techniques that can adapt to new situations and that can exploit coordination, task variations and perception uncertainties. The chapter presents an overview of the research lines of the Robot learning & Interaction group at the Idiap Research Institute. It analyses the barriers that still need to be overcome to move these techniques out of the labs, by complying with real industrial application constraints. It also proposes a variety of research directions for which further investigations would be required to overcome these barriers. |
| Keywords: | learning from demonstration, robotics |
| Projects |
Idiap |
| Authors | |
| Editors | |
| Added by: | [UNK] |
| Total mark: | 0 |
|
Attachments
|
|
|
Notes
|
|
|
|
|