Agent-based control for fuzzy behavior programming in robotic excavation

This paper discusses the concept, formulation, and implementation of the agent-based control for fuzzy behavior programming in robotic excavation. Petri net transducers are introduced to describe excavation control agent coordination and specification, while fuzzy control rules are used to implement primitive motions. A prototype laboratory excavation system is built with PUMA robotic manipulators and a force/torque sensor. Extensive experiments have been conducted and the results have demonstrated that the proposed control method is capable of continuously adapting and replanning its actions based on sensory feedback, and completing its excavation tasks in dynamic and unstructured environments.