On the manipulation of articulated objects in human-robot cooperation scenarios

Articulated and flexible objects constitute a challenge for robot manipulation tasks, but are present in different real-world settings, including home andindustrial environments. Approaches to the manipulation of such objects employ ad hoc strategies to sequence and perform actions on them depending on their physical or geometrical features, and on a priori target object configurations, whereas principled strategies to sequence basic manipulation actions for these objects have not been fully explored in the literature. In this paper, we propose a novel action planning and execution framework for the manipulation of articulated objects, which (i) employs action planning to sequence a set of actions leading to a target articulated object configuration,and (ii) allows humans to collaboratively carry out the plan with the robot, also interrupting its execution if needed. The framework adopts a formally defined representation of articulated objects. A link exists between the way articulated objects are perceived by the robot, how they are formally represented in the action planning and execution framework, and the complexity of the planning process. Results related to planning performance, and examples with a Baxter dualarm manipulator operating on articulated objects with humans are shown.