Automatically acquiring knowledge by digital maps in artificial intelligence planning techniques

Any ES research, in particular planner ES research, shows that knowledge acquisition is a bottleneck when building up the ES prototypes. From this viewpoint, the possibility of automatically acquiring knowledge for ES, at least with reference to special themes and problems, may be seen as constituting an interesting line of research. Inductive and EBL methodologies, founded on quantitative and qualitative knowledge, are the main contributions that the ML (a branch of AI) makes to the above mentioned goal. Both approaches, even if very little investigated in the planning domain, seem to invite further research. We refer in particular to the potentials of automatically learning by maps. Relating to the research on planner ES prototypes for urban environmental control, this paper intends to set up a preliminary discussion on the perspectives offered by the automatic interpretation of key variables in order to make elementary inferences concerning urban typologies and situations