UPDATE: July 2007
Re-entrant PP failed to yield the initial gains when long runs with forgetting were tried. That project has been put on one side until new ideas make it worth proceeding.
The current project, called PPP (“triple P”), adds a special plan memory to PP in the spirit of Bernard Baars' well-known “Global Workspace” ( In the Theater of Consciousness , 1997) and David Rosenthal's equally well-known “Higher Order Thought” theory ( Consciousness and Mind, 2005). My goal is a robot that “knows what it is doing” and satisfies my “yammy-5” definition ( Associative Learning for a Robot Intelligence, 1998):
Definition-5. A robot will be described as yammy-5 when its behaviour in an open environment is most efficiently predicted in terms of
(i) its own intentions,
(ii) its knowledge of the intentions of other purposeful entities,
(iii) its knowledge of their knowledge of its intentions,
(iv) its ability to predict its own and other entities' future intentions, and
(v) the answers it gives to its own questions about its own intentions.
The plan memory has been added to PP and plan-monitoring is being programmed. The Squash-Pop world allows interaction between a robot with a PP brain and another robot and food objects.