Work at SourceForge, help us to make it a better place! We have an immediate need for a Support Technician in our San Francisco or Denver office.
I've been working on research regarding adding certain plan/intention maintenance functionality to BDI agents - this entails adding a 'maintenance phase' which occurs after event handling/belief update, and before the consideration of new desires/options.
However, I've been struggling to find/decide upon a suitable architecture to implement a proof of concept; which is a roundabout way of asking whether this possible within the Jadex architecture?
in Jadex there is no deliberation cycle in which consistently typical steps are executed. We found this model forces too strict. In Jadex BDIV2 the BDI behaviour is controlled via a RETE rule engine, i.e. all meta actions (like creating a goal, executing a plan, deliberate about a goal) are specified as rules. Extensions also can be integrated by adding new rules to the underlying rule engine.