HYCAS'09

[edit] International Workshop on Hybrid Control of Autonomous Systems

Integrating Learning, Deliberation and Reactive Control

About this Workshop

High-level control for Autonomous Systems (e.g. robots) is concerned with selecting the next action the system should perform. In particular this means that the system must be endowed with algorithms or schemes to take the next step towards its mission goal. The known paradigms for this action selection problem are learning, deliberation, reactive control schemes or combinations of these schemes, i.e. hybrid approaches. Learning has been applied successfully to many robotics tasks. Most of the work is related to learning certain basic behaviours or skills. Examples where the high-level control strategy of robots (or agents) were successfully learned are rare. The deliberative approach for decision making of autonomous systems was successfully treated in research on Artificial Intelligence, following a top-down approach, which has severe limitations in real applications. In the reactive control paradigm the idea is that, with a combination of purely reactive action selection schemes, intelligent and goal-directed behaviours emerge, which can be seen as a bottom-up approach. These different paradigms have been known for over two decades, and in fact, in today's applications, often combinations of learning, deliberation and reactive control are used. Usually these combinations are used in an ad-hoc or even unconscious fashion. Although there is a number of proposed architectures and huge body of literature, the issue of combining learning, reactive and deliberative control never has been intensively investigated. With this workshop we wish to bring together researchers from different areas who concentrate on combinations of learning, planning, and/or reactive schemes for decision making and the control of autonomous systems. The workshop is open to all members of the AI and Robotics community. We would specifically like to encourage students to participate.