Theoretical and experimental results on the goal-plan tree problem
Patricia H. Shaw, Berndt Farwer, Rafael H. Bordini · Adaptive Agents and Multi-Agents Systems · 2008
Agents programmed in BDI-inspired languages have goals to achieve and a library of plans that can be used to achieve them, typically requiring further goals to be adopted. This is most naturally represented by a structure that has been called a Goal-Plan Tree. One of the uses of such structure is in agent deliberation (in particular, deciding whether to commit to achieving a certain goal or not). This paper presents new experimental results combining various types of goal-plan tree reasoning from the literature.