Jerry B. Weinberg, Ph. D.
Acting Associate Provost for Research and
Dean The Graduate School

This project demonstrates that an inexpensive platform is capable of being used to teach a deliberative robotic control that includes navigation and planning. Navigation incorporates determining a goal location, planning a path to get there, maintaining knowledge of where the robot has been, and maintaining present location within the world representation. Planning involves taking knowledge that the robot has acquired and using it to develop a sequence of actions that achieve a goal. To enable a more direct compare-and-contrast of a deliberative architecture?s performance versus a reactive architecture?s performance, two robots were developed. The two LEGO Mindstorms robots were built using standard pieces with the exact same physical properties. One construct housed an RCX with a deliberative software architecture, the other and RCX with reactive architecture programming. Once development and testing of the two platforms was complete, the deliberative robot design was used to teach robotics concepts of planning and navigation in an upper-division class on artificial intelligence. To maximize the applicability of the results of this project to Southern Illinois University Edwardsville and other universities, the project makes use of the LEGO Mindstorms set, a commercially available suite of LEGO rotation, light, and touch sensors, and C-like programming environments that can be obtained by educational institutions free of charge.

Implementation of a Deliberative Robot Control Architecture on an Inexpensive Robot Platform

Gary R. Mayer

Abstract