Robotics offer an interesting challenge for integration of a rich set of different technologies into operational systems. Construction of such systems requires use of techniques for mechanics, control engineering, perception, software engineering, and artificial intelligence. The need for such diverse expertise has made it difficult for individual groups / companies to provide the necessary critical mass that allow construction of fully operational systems. In addition fundamental problems remains to be solved before flexible and robust systems can be deployed in realistic setting (at a reasonable cost).

To facilitate progress in these fundamental issues a series of Dagstuhl seminars on Modeling of Sensor Based Intelligent Robot Systems has been organized.

This includes the seminars

• Environmental Modeling and Motion Planning for Autonomous Robots, Organized by: Bunke, Kanade and Noltemeier,
• Modelling and Planning for Sensor Based Intelligent Robot Systems, Organized by: Bolles, Bunke, and Noltemeier and
• Modelling and Planning for Sensor-Based Intelligent Robot Systems, Organized by: Bolles, Bunke, Christensen and Noltemeier.

These workshops have demonstrated that robotics gradually to becoming mature as a few systems have been deployed in industrial settings. It has further been demonstrated that perception and systems integration remains the two the major obstacles to construction of systems. Recent research has also demonstrated that the issue of human-robot interaction is a new and interesting topic that call for methods that are different from those studied in traditional human-computer interaction. It has further become clear that it is difficult to program these systems to carry out a rich set of tasks, which call for methods from machine learning to enable new ways to instruct the robots to carry out specific tasks.

The year 2000 seminar will consequently focus on the topics of:

• Sensing for autonomous operation
• Planning for dynamic environments
• System integration
• Human-Robot Interaction
• Learning and Knowledge Acquisition

Each day a well known expert from each of these areas will provide a key-note presentation, which will be followed by regular papers. At the end of the day a panel will be organized to assess state of the art and identify fundamental new problems.