AND/OR net representation for robotic task sequence planning

摘要

The paper describes a framework for task sequence planning for angeneralized robotic work cell. The AND/OR net provides a compact,ndistributed, domain-specific representation of geometric configurationsnof parts and devices in the work cell. The approach maintains ancorrespondence from geometric state information to task and motion plansnand on-line discrete-event control that is not available in traditionalnaction-based planners. The feasibility criteria for each AND/OR netntransition guide the geometric reasoning required in the planning ofnfeasible sequences. The resulting search space for plans is often muchnsmaller (due to explicit representation of geometric constraints) thannthe state space of an action-based task planner. For purposes ofnanalysis, the AND/OR net is mapped into a Petri net and the resultingnPetri net is shown to be bounded and have guaranteed properties ofnliveness, safeness, and reversibility. In this form, the AND/OR net maynbe viewed as a Petri net synthesis tool in which the resulting Petri netnrepresentation may be used for on-line scheduling and control of thensystem