Unmanned aerial vehicles (UAV's) have received an increasing amount of
attention in recent literature. Investigations and developments over the past
years have been focused mostly in military applications. Nowadays, it is
possible to build lightweight and tiny UAVs, with excellent capacities to develop
and deploy civilian applications in benefit of the citizenship and society.
Motivated by recent advances in intelligent systems, cooperative control and
studies on large groups of unmanned autonomous vehicles or UAVs, this
project identifies some problems and key points in order to create a
decentralized mission planner for controlling an heterogeneous group of UAVs.
Based on this gathered experience, technological and scientific research
focused on the use and exploitation of multi-agent systems for mission
planning and control of a UAVs network is presented. In the first sections, the
problem statement is settled. After reviewing multi-agent systems, the main
environmental model is defined and a reduced version addressed for system
identification is fixed. Next, the MAS concept is used for the mission planning
task. Some results are provided from this approach. Finally, conclusions and
recommendations for further research are presented.
This project has been done in collaboration with the Aerospace Research
Technology Centre (CTAE) which investing research effort in UAVs and their
applications.
