T9 - FULL DAY ================================================================= Robotics for Beginners; Using Robot Kits to Build Embodied Agents E. Sklar, S. Parsons MOTIVATION Robots have always been prototypical agents; they are situated in the physical world and have all the classic problems of noisy sensors and imprecise actuators. However, until recently, to be a robotics researcher you had to be a "real" engineer with an advanced degree in electrical and/or mechanical engineering and a big lab filled with computers, soldering irons, a drill press and a lathe. Today, you only have to be an eight-year-old child with a PC and a penchant for playing with LEGO. Just as the trend in computer hardware has evolved over the last 50 years from immense machines that took up entire rooms to powerful microprocessors that literally fit in the palm of your hand, robots have gotten smaller, faster and cheaper. Through kits like the LEGO Mindstorms, agents researchers have access to robots much more easily than before, both for teaching and research. The kits are being used in introductory robotics classes, even in departments where there is no robot research lab. As well, a few faculty members are bringing the kits into classes that are not specifically about robots, but as a means for giving students a hands-on platform on which to experiment with the some of the abstract concepts they are learning -- for example, in introductory programming, artificial intelligence and multi-agent systems. The purpose of this tutorial is to show how the LEGO Mindstorms kit can be used as a means for exploring agents topics, such as individual agent control architectures, swarm-like group behaviours and agent communication mechanisms. The full-day session will begin with an overview of the basic concepts of autonomous agents and autonomous robotics. Then attendees will be given instruction on the use of the hardware and software -- a freeware C-like programming language (called Not-Quite C or NQC), and will carry out some simple exercises to get hands-on experience with the robotics kits (provided by the presenters) and how to program them. This will be followed by presentation of progressively more complex examples of agent behaviours implemented on the robotics kits. Each example will be accompanied by an exercise in which the tutorial attendees implement the behaviour themselves, either from scratch or based on code provided by the presenters. The final segment of the course will contain examples of the use of robotcs kits in undergraduate curriculum, based on courses developed by the presenters. The session will close with discussion of possible uses of the kits based on the attendees' needs. Through attending this tutorial, an attendee will: * understand the basic concepts of autonomous agents and autonomous robotics; * be able to build a LEGO Mindstorms robot and program it using Not-Quite C; * gain experience of implementing simple techniques from robotics on the LEGO Mindstorms platform; * understand how robotics kits can be used as a basis for building embodied agents; and * learn how robotics kits can be used effectively in undergraduate classrooms. =========================================================================== IMPORTANT NOTE Although the presenters plan to use LEGO Mindstorms kits in this tutorial, they hold no brief from LEGO, nor do they endorse the product. The LEGO kit is one of a number of similar kits on the market, all of which will be discussed in the tutorial. LEGO kits will be used for the tutorial simply because the presenters happen to have enough such kits for attendees to use. ======================================================================= INTENDED AUDIENCE The tutorial is intended for any delegates who have some basic knowledge of aspects of artificial intelligence, and knowledge of an imperative programming language. ========================================================================= BACKGROUND KNOWLEDGE REQUIRES Although not necessary, it will be helpful if attendees bring their own laptop, running Linux or Windows. Registered attendees will be contacted by the presenters with information on downloading freeware prior to the tutorial. ============================================================================ DETAILED OUTLINE * Presentation: Autonomous agents and autonomous robotics (1 hour). * Presentation: NQC and the Lego Mindstorms. (0.5 hour). * Exercise 1: Build and program a simple line-following robot (includes testing robots developed by participants, competition over an oval track, and discussion of outcome). (1 hour). * Presentation: Behaviour-based robotics. (0.5 hour). * Exercise 2: Build and program a behaviour-based robot that can follow a line, move around obstacles it hits and head towards a light (includes a competition over an obstacle course, and discussion of outcome). (1.5 hours). * Presentation: BDI Architecture (0.5 hour). * Exercise 3: Build and program a BDI-controlled robot that can play a simple game of soccer using a ball that emits infra-red light (includes a small soccer tournament and discussion of outcome). (1.5 hours). * Presentation: Using robotics kits in undergraduate education. Case studies include courses in robotics, introductory programming, artificial intelligence and multi-agent systems. Description of alternative robotics kits. (1 hour). * Discussion: Applying the ideas and techniques presented to attendees' teaching and research projects. (0.5 hour). ==================================================================== BIOGRAPHY OF PRESENTERS Elizabeth Sklar Dept of Computer Science, Columbia University, 1214 Amsterdam Avenue, MC 0401, New York, NY 10027 USA tel: +1 212 939 7021 fax: +1 212 666 0140 email: sklar@cs.columbia.edu Elizabeth Sklar is an Assistant Professor in the Computer Science department at Columbia University. She earned her PhD in 2000 from Brandeis University for developing a framework for an interactive, agent-based on-line system applied to education. She is the international chairperson of RoboCupJunior and has been studying the educational value of this initiative since its inception in 2000. One aspect of this research entails evaluating robotic kits for use in classrooms. Dr Sklar works with middle and high school teachers to help them integrate technology effectively into their curriculum. Dr Sklar is a member of the Executive Committee of the RoboCup Federation. She is also the co-chair of a new ACM project for training high school teachers to teach Java. Her research lab studies learning environments in which humans and agents -- simulated and embodied -- interact and co-evolve. Simon Parsons Dept of Computer and Information Science, Brooklyn College, City University of New York, 2900 Bedford Avenue, Brooklyn, NY 11210 USA tel: +1 718 951 4139 fax: +1 718 951 4842 email: parsons@sci.brooklyn.cuny.edu Simon Parsons is a Associate Professor in the Department of Computer and Information Science at Brooklyn College, City University of New York. Dr Parsons received his PhD from the University of London in 1993 and subsequently worked at the Imperial Cancer Research Fund, the University of London, the University of Liverpool, and MIT. During the last year Dr Parsons has been using the LEGO Midstorms kit as a vehicle for teaching undergraduate courses on artificial intelligence. Dr Parsons has written over one hundred papers on techniques supporting decision making in intelligent systems and co-edited three collections and written a monograph on the same subject. Dr Parsons is the editor of the Knowledge Engineering Review, published by Cambridge University Press, and was the recipient of a 1998 Younger Engineers Achievement Medal, awarded by the IEE, for his work in informatics.