Invited Talks

ICAPS/AAMAS Invited Talk

Robotic Agents for Disaster Response Robotics

Daniele Nardi, Università di Roma "La Sapienza"

Disaster Response Robotics is a challenging domain, where the need for intelligent robotic agents (as opposed to just robots) is motivated both from technical considerations and from a practical application perspective. After providing arguments to support the above claim, I will briefly overview the state of the art in the field. Then, I will address some of the research approaches that we have developed at Sapienza Univ. of Rome, also within a collaboration with the Italian Firemen Department. Specifically, I will present some research in Distributed Situation Assessment, Action Planning and Monitoring, Context-based Design of intelligent robotic agents, Multi-robot Teams for disaster response robotics and Performance Evaluation Metrics for intelligent robotic agents. Throughout the discussion of past research, I will try to focus on several open challenges that need to be solved in order to provide effective solutions for Disaster Response Robotics.

KR/AAMAS Invited Talk

Great Moments in KR: The 1984 Complexity Convergence

Ron Brachman, Yahoo! and Hector Levesque, University of Toronto

The 1970's were a fertile and exciting time for Artificial Intelligence. This was especially true in the area of Knowledge Representation, where numerous novel languages and systems were created and debated, and a burgeoning set of AI applications were supported by frames, semantic networks, production rules, and other idiosyncratic KR schemes. But the issues under debate were often vague, and intuition and implementation-based arguments reigned, with little or no formal basis for discussion. Fortunately, in the late '70's and early '80's, out of this energetic but murky environment emerged several important lines of thought that promised to put elements of the field on a firmer foundation. By 1984 these threads had been developed enough that they could be knitted together, and out of this convergence a new kind of formal analysis of KR systems became possible. We look at the birth and evolution of several key ideas and how they came together to allow us to draw some interesting conclusions about the computational complexity of core inferences in a mainstream form of KR. We also make some observations about the aftermath, and how that moment in the history of the field seemed to mark a sea change in approaches to Knowledge Representation.