Plan representations for picking up trash

This paper has not already been accepted by and is not currently under review for a journal or another conference. Nor will it be submitted for such during IJCAI's review period. Abstract In this paper we describe the skills and RAPs used by the University of Chicago entry in the 1994 AAAI robot competition. The robot was controlled using the Animate Agent Architecture and the skills and RAPs presented constitute the knowledge represented within that architecture for performing the trash gathering task. Along with describing the knowledge represented for the task, this paper discusses some of the lessons learned in extending the skills and RAPs to be more genericc so they can be used for a variety of other tasks. In particular, realistic reactive plans must contain signiicant concurrency. Furthermore, reasonable strategies for carrying out ordinary tasks often require explicitly unordered tasks to take advantage of opportunities that might arise for achieving subgoals in alternate orders.