Skip to main content

A realistic testbed for Multi-Agent Cooperation: the Robot Rescue Application

Dr. Arnaud Visser ( Univertiteit van Amsterdam )
Cooperation between autonomous agents is an active field of research, which has many applications. Cooperation within a team of robots has its own challenges. Before decisions can be made, a joint world model has to be made based on incomplete and inaccurate measurements from distributed locations. Based on the part of the joint world available to a single robot the possible actions have to be coordinated inside the team to achieve common goals. When these challenges are solved, the team can greatly benefit from the cooperation.??This potential is demonstrated with the use of mobile robots for search and rescue missions after a disaster. During rescue operations for disaster mitigation, cooperation is a must. This requirement is due to the structural diversity of disaster areas, variety of evidence of human presence that sensors can perceive, and the necessity to quickly and reliably examine targeted regions. To survive in the harsh environmental conditions of a rescue mission, the robots have to combine many state-of-the-art Artificial Intelligence algorithms. Rescue missions are the ultimate benchmark for those algorithms, which is demonstrated at the yearly RoboCup Rescue competition. The talk will focus on the progress recently made in this field.??This is joint work with J. de Hoog and Dr. S. Cameron

 

 

Share this: