A Big but Bridgeable Gap: From Deliberative Planning to Real-Time Execution and Learning

Manuela Veloso

Problem solving is a complicated intelligent process that includes planning, execution, and necessarily learning. The Prodigy problem solving architecture contributed multiple successful approaches to the integration of planning and learning. Execution offers a variety of additional challenges to problem solving, including real-time decision making, uncertainty in state and action modeling, and possibly multiple executors. I have always researched on problem solving, both within the Prodigy architecture and also on complete teams of robot soccer executors. In this talk, I will discuss the existing gaps between classical planning, execution, and learning. I will also present our efforts and contributions to bridge that gap.

This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.