No Stylesheet Loaded

Could not load Kickstrap.There are several common reasons for this error.

Maria Gini, Department of Computer Science and Engineering, University of Minnesota, USA

Exploiting spatial locality and heterogeneity of agents for search and rescue teamwork

When a large-scale disaster strikes, it is critical for rescuers to coordinate their work effectively despite time pressure and limited knowledge of the situation. We propose different types of team structures and compare their performance using the RoboCup Rescue simulation as a testbed. We measure the effectiveness of the teaming strategies using task specific metrics (such as time to save people, and number of buildings destroyed by fire), and system level metrics (such as computation time, and scalability to number of agents). Our results support the hypothesis that teamwork improves performance, and that more specialized and knowledge-rich teaming arrangements perform better. [SLIDES PDF]

Maria Gini is a Distinguished Professor of the College of Science and Engineering at the University of Minnesota. Her research focuses on multi-robot and multi-agent systems. She leads active research efforts in allocation of tasks that have time and precedence constraints, teamwork, exploration of unknown environments with robots, decision making for autonomous agents, and learning of opponent behaviors. She is a Fellow of AAAI and a Distinguished Scientist of ACM. She is on the editorial board of multiple journals including Artificial Intelligence; Autonomous Agents and Multi-Agent Systems; Robotics and Autonomous Systems.

Markus Kuchler, Ericsson Response, Sweden

Ericsson Response

Ericsson Response - A group-wide employee initiative within the Ericsson group for ICT support in humanitarian emergencies. [SLIDES PDF]

Markus Kuchler is the Program Director for Ericsson Response. Before his current assignment Markus spent 10 years in different management positions within the Ericsson global Corporate Security organization. Prior to that he worked within law enforcement as a criminal investigator of juvenile and computer crimes. He holds a law degree and has the rank of major (reserve) within the Swedish Armed forces.

Simon Lacroix, LAAS, France

Autonomous teams of aerial and ground robots in search and rescue missions

The deployement of coordinated aerial and ground robots yields numerous operational benefits, that are mainly due to the complementarities of both kinds of systems regarding their sensing and acting abilities. From a robotic standpoint, such systems of systems also allow the development of cooperative schemes and synergies, and raise several research issues on the usual perception and decision functionalities.

The talk will analyse the required functionalities for the deployment of teams of autonomous aerial and terrestrial robots over large scales of space and time, under various conditions. It will focus on the importance of environment representations, which are the main information on which autonomous cooperation schemes can be defined. Even when a priori available, environment models need to be updated and/or refined by the robots: that fact that the perception processes are distributed among heterogeneous robots challenges the sensor fusion schemes developed for single robots, and in particular for localization, which plays a central role in the management of the environment models. The presentation will be illustrated with various results, and draw some research perspectives for the autonomy of robot teams in search and rescue applications. [SLIDES PDF]

Short Bio: Simon Lacroix graduated from Ecole Centrale de Paris in 1990. He then prepared a PhD at LAAS, on autonomous navigation for planetary rovers. After a one-year postDoc in the Center for Intelligent Machines of McGill University, Montreal, he joined LAAS as a permanent research scientist in 1996. He currently animates the field robotics activities of the laboratory, and his research are focused on the deployment of teams of multiple heterogeneous autonomous robots for exploration, surveillance or intervention missions. His main interests are perception and navigation for autonomous aerial and terrestrial robots (environment perception and modeling, localization, perception control and autonomous navigation strategies), and decisional processes for autonomous cooperation within multi-robot teams.

Claudio Melchiorri, DEI, Dept. of Electrical, Electronic and Information Engineering University of Bologna and CAI - Italian Alpine Club

Application of Robotic Technologies to Mountain Rescue

Mountain search and rescue operations are challenging and often quite dangerous for human operators because of the many difficulties related to the harsh terrain (steep rock, snow, vegetation, ...), objective dangers, different type of weather and environmental conditions (fog, rain, wind, night, cold, etc.). In addition, another difficulty is related to the quantity and the number of different operations, that ranges from the search of buried persons in avalanches to the search of missing people in a wide region, or to the rescue of a rock climber on a vertical face of a mountain. In this type of operations, robotic technology may have an important role in supporting human rescuers in many aspects of their activities. [SLIDES PDF]

In the talk, mountain search and rescue operations will be described, reporting statistics coming from the Alpine mountain chain and focusing on some of the most challenging missions. Moreover, some ideas about the use of robotic devices and related technologies in this type of operations will be given, on the basis of the activities and goals of the EU project SHERPA.

Paul Scerri, Robotics Institute, Carnegie Mellon University (CMU), USA

Low-Cost Robotic Airboats for Information Collection

Collecting information about large bodies of water, including lakes used for fish farming and flood waters, is difficult, dangerous and expensive to do manually, but it is a perfect job for robots. Teams of low-cost robotic boats can be used to get spatially and temporally dense data about water, including looking for disease outbreaks or people trapped by flood waters. This talk will describe some progress on building teams of such boats, with a particular focus on the advantages to getting out into the field as early as possible in the development process and continually getting into the field over the course of the development. [SLIDES PDF]

Short Bio: Paul Scerri is an Associate Research Professor at Carnegie Mellon University's Robotics Institute. His research looks at how to coordinate many robots or intelligent agents to perform very complex tasks. He received his PhD from Linkoping University, did his postdoc at the University of Southern California and has been at Carnegie Mellon University since 2003. Paul is on the editorial board of several journals, including the JAAMAS and will be Program Chair for AAMAS 2014 in Paris.