Projects / Programmes
Ciljno usmerjena sinteza robotskih operacij na osnovi baze vzorčnih gibanj (Slovene)
| Code |
Science |
Field |
Subfield |
| 2.10.04 |
Engineering sciences and technologies |
Manufacturing technologies and systems |
Robotics |
| Code |
Science |
Field |
| 2.11 |
Engineering and Technology |
Other engineering and technologies |
Organisations (1)
, Researchers (8)
0106 Jožef Stefan Institute
| no. |
Code |
Name and surname |
Research area |
Role |
Period |
No. of publicationsNo. of publications |
| 1. |
33646 |
PhD Miha Deniša |
Manufacturing technologies and systems |
Technical associate |
2011 - 2012 |
56 |
| 2. |
25638 |
PhD Andrej Gams |
Manufacturing technologies and systems |
Researcher |
2009 - 2010 |
253 |
| 3. |
00118 |
PhD Bojan Nemec |
Systems and cybernetics |
Researcher |
2009 - 2011 |
299 |
| 4. |
21549 |
PhD Damir Omrčen |
Manufacturing technologies and systems |
Researcher |
2009 - 2010 |
51 |
| 5. |
06654 |
Jožef Opeka |
|
Technical associate |
2009 - 2010 |
11 |
| 6. |
34573 |
Luka Peternel |
Manufacturing technologies and systems |
Researcher |
2011 |
39 |
| 7. |
32871 |
Goran Škorja |
Manufacturing technologies and systems |
Researcher |
2011 |
7 |
| 8. |
11772 |
PhD Aleš Ude |
Manufacturing technologies and systems |
Head |
2009 - 2012 |
495 |
Significance for science
One of the biggest challenges of modern robotics is to enable general purpose robots to autonomously act in natural environments such as homes. The development of such systems will enable the transfer of robot technologies from rather static industrial environments to more unpredictable and dynamic natural environments, thus significantly enlarging the robotics market. In this project we developed a new robot learning paradigm methodology that enables the robots to
- Acquire initial sensorimotor knowledge to solve a given task in different configurations of the external world. We used methods such as the observation of human motion and kinesthetic guiding to enable quick transfer of task knowledge to a robot.
- Transfer of human motion to the robot's kinematics and dynamics.
- Generalization of the acquired action knowledge to new situations, thereby enabling the synthesis of novel robot trajectories.
- The application of task parametrization to accelerate reinforcement learning.
- Automatic identification of parameters that describe the desired task.
This way we contributed to the development of cognitive robotic systems.
Significance for the country
Cognitive robotics is one of the important topics of the seventh framework programme of the European Union. With this project we contributed to the development of cognitive robotics in Slovenia. Our programme group has recently acquired four new FP7 projects in the area of "Cognitive systems and robotics". This way we significantly increased the research activity in this area in Slovenia and contributed to the participation of Slovenia in the EU seventh framework programme.
Results of our projects provide some of the necessary foundations for the development of general-purpose robotic systems that can help people in natural environments, for example their homes. These type of robots could significantly contribute to improving the quality of life, especially for older people. If robotics prolonged the time that people can live at their homes independently and without the help of other people just for a few months, it would significantly contribute to solving the problems of the aging society. It is clear that the development of such robotic systems still requires large investment and integration in European and global research activities. It is therefore important that Slovenia remains at the forefront of research in this field.
Most important scientific results
Annual report
2009,
2010,
2011,
final report
Most important socioeconomically and culturally relevant results
Annual report
2009,
2010,
2011,
final report