Projects / Programmes
Modelling and control of multiagent systems
| Code |
Science |
Field |
Subfield |
| 2.06.00 |
Engineering sciences and technologies |
Systems and cybernetics |
|
| Code |
Science |
Field |
| T125 |
Technological sciences |
Automation, robotics, control engineering |
| T111 |
Technological sciences |
Imaging, image processing |
| P175 |
Natural sciences and mathematics |
Informatics, systems theory |
| P176 |
Natural sciences and mathematics |
Artificial intelligence |
multiagent systems, computer vision, modelling, control strategy
Researchers (2)
| no. |
Code |
Name and surname |
Research area |
Role |
Period |
No. of publicationsNo. of publications |
| 1. |
01952 |
PhD Rihard Karba |
Systems and cybernetics |
Head |
2004 - 2006 |
620 |
| 2. |
20181 |
PhD Gregor Klančar |
Systems and cybernetics |
Researcher |
2004 - 2006 |
318 |
Organisations (1)
Abstract
Multiagent systems perform a part of well known artificial intelligence area. They offer the effective possibility of complex tasks execution by the aid of multiple agents, which cooperate in attaining the prescribed goal. This young area is very popular and represents the object of intense research. The agent is frequently an autonomous mobile robot which must have the possibilities of environment perception, of thinking and decision making, of the decision realization and of influencing the environment. The application areas are in some industries (servicing of specific industrial cells, warehouses, etc.), in some dangerous and/or inapproachable environments as well as in education and entertainment. The domain has a strong interdisciplinary character, including areas such as: modelling, simulation, control, system theory, robotics, computer vision, artificial intelligence and communications. Worldwide recognized aid for studying multiagent systems problems is the pilot structure in the shape of robot soccer, where two teams of mobile robots play in the disturbed and competitive environment.
The investigation will be focused in the following directions:
- study of mobile robots optimal control in their way to the target considering robot dynamics, environment limitations and obstacles avoidance,
- study of selflearning agents, which increases their adaptability to the environment,
- upgrading the realistic simulator which facilitates the study of different approaches,
- study of computer vision in the sense of combining the information from camera, sensors on the agent and control algorithm which is known as visual servoing,
- study of different control strategies in the sense of selforganization and cooperation of agents.
All developed approaches will be tested on the simulator as well as on the robot soccer rig with the modifications in the sense of simulating some situations in the industrial applications.
