Automated Planning Encodings for the Manipulation of Articulated Objects in 3D with Gravity

Abstract

The manipulation of articulated objects plays an important role in real-world robot tasks, both in home and industrial environments. A lot of attention has been devoted to the development of ad hoc approaches and algorithms for generating the sequence of movements the robot has to perform in order to manipulate the object. Such approaches can hardly generalise on different settings, and are usually focused on 2D manipulations. In this paper we introduce a set of PDDL+ formulations for performing automated manipulation of articulated objects in a three-dimensional workspace by a dual-arm robot. Presented formulations differ in terms of how gravity is modelled, considering different trade-offs between modelling accuracy and planning performance, and between human-readability and parsability by planners. Our experimental analysis compares the formulations on a range of domain-independent planners, that aim at generating plans for allowing a dual-arm robot to manipulate articulated objects of different sizes. Validation is performed in simulation on a Baxter robot.