Discrete Event Plan Representation

Using Discrete Event and Hybrid systems to represent task plans brings exciting new possibilities to include commitment and synchroniztion mechanisms between cooperating robots, and to develop plan qualititative (e.g., deadlocks, liveness) and quantitative (e.g., time taken by a plane, probability of success of a plan) analysis and synthesis procedures. Some work in this direction has been done - check the paper "Stochastic Discrete Event Model of a Multi-Robot Team Playing an Adversarial Game".

Relational Behaviors

Relational behaviors are interesting to us if they are defined under some formal framework, such that whatever we do in soccer with them can be used as a paradigm for relational behaviors in more general application problems. This was the procedure followed in the pass implementation described in RoboCup2004 Symposium paper "Formulation and Implementation of Relational Behaviours for Multi-Robot Cooperative Systems". This was work done by our Dutch student Bob van der Vecht.

The work is being continued by further research on implementing Bob's method using Petri nets.