Integrated Robot Task and Motion Planning in the Now

Abstract

Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. This paper provides an approach to integrating geometric motion planning with logical task planning for long-horizon tasks in domains with many objects. We propose a tight integration between the logical and geometric aspects of planning. We use a logical representation which includes entities that refer to poses, grasps, paths and regions, without the need for a priori discretization. Given this representation and some simple mechanisms for geometric inference,we characterize the pre-conditions and effects of robot actions in terms of these logical entities. We then reason about the interaction of the geometric and non-geometric aspects of our domains using the general-purpose mechanism of goal regression (also known as pre-image backchaining). We propose an aggressive mechanism for temporal hierarchical decomposition, which postpones the pre-conditions of actions to create an abstraction hierarchy that both limits the lengths of plans that need to be generated and limits the set of objects relevant to each plan. We describe an implementation of this planning method and demonstrate it in a simulated kitchen environment in which it solves problems that require approximately 100 individual pick or place operations for moving multiple objects in a complex domain. This paper provides an approach to integrating geometric motion planning with logical task planning for long-horizon tasks in domains with many objects. We propose a tight integration between the logical and geometric aspects of planning. We use a logical representation which includes entities that refer to poses, grasps, paths and regions, without the need for a priori discretization. Given this representation and some simple mechanisms for geometric inference, we characterize the pre-conditions and effects of robot actions in terms of these logical entities.We then reason about the interaction of the geometric and non-geometric aspects of our domains using the general-purpose mechanism of goal regression (also known as pre-image backchaining). We propose an aggressive mechanism for temporal hierarchical decomposition, which postpones the pre-conditions of actions to create an abstraction hierarchy that both limits the lengths of plans that need t...

Integrated task and motion planning in belief space

“…We have built upon this domain to make the robot search for objects in unknown cupboards [Wong et al, 2013], and to solve simple surveillance problems. In the completely observable case [Kaelbling and Lozano-Pérez, 2012a], we have built upon a basic pick-and-place capability to encode a kitchen domain involving cooking and washing dishes (in a somewhat abstract formulation). The additional operators to extend to the kitchen domain were relatively easy to construct.…”

“…There have been several recent approaches to integrated task and motion planning [Cambon et al, 2009, Plaku and Hager, 2010, Marthi et al, 2010, Dornhege et al, 2009 with the potential of scaling to mobile manipulation problems; we review this related work in detail in [Kaelbling and Lozano-Pérez, 2012a]. However, none of these approaches addresses uncertainty directly.…”

“…Our approach to hierarchy is applied to determinized models, and so it is more closely related to the deterministic hierarchical planning literature (the relationship is outlined by Kaelbling and Lozano-Pérez [2012a]). …”

“…We extend the basic hpn algorithm [Kaelbling and Lozano-Pérez, 2012a], to apply to domains with outcome uncertainty by retaining the same depth-first planning and execution architecture, but additionally monitoring the effects of actions to ensure that the action being currently selected is the first step in a plan that has positive probability of achieving the goal, and that obviously unnecessary actions are not taken.…”

“…As described in detail in [Kaelbling and Lozano-Pérez, 2012a], we create abstract versions of operators by postponing consideration of some of the preconditions. Assume that we have an abstract version of Op1 that has postponed the precondition A.…”