Modular task and motion planning in belief space

Hadfield-Menell, Dylan; Groshev, Edward; Chitnis, Rohan; Abbeel, Pieter

doi:10.1109/iros.2015.7354079

Cited by 45 publications

(44 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The current approach is benchmarked against two approaches: 1) an ontological physics-based motion planner, introduced in [30], that enhances KPIECE for physics-based motion planning (it will be referred as o-KPIECE); 2) a task and motion planning approach for grasping in clutter proposed in [2], [3]. We run 30 simulation for each scenario.…”

Section: B Benchmarkingmentioning

confidence: 99%

See 1 more Smart Citation

Randomized Physics-Based Motion Planning for Grasping in Cluttered and Uncertain Environments

Muhayyuddin

Moll

Kavraki

et al. 2018

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

Abstract-Planning motions to grasp an object in cluttered and uncertain environments is a challenging task, particularly when a collision-free trajectory does not exist and objects obstructing the way are required to be carefully grasped and moved out. This paper takes a different approach and proposes to address this problem by using a randomized physics-based motion planner that permits robot-object and object-object interactions. The main idea is to avoid an explicit high-level reasoning of the task by providing the motion planner with a physics engine to evaluate possible complex multi-body dynamical interactions. The approach is able to solve the problem in complex scenarios, also considering uncertainty in the objects' pose and in the contact dynamics. The work enhances the state validity checker, the control sampler and the tree exploration strategy of a kinodynamic motion planner called KPIECE. The enhanced algorithm, called p-KPIECE, has been validated in simulation and with real experiments. The results have been compared with an ontological physics-based motion planner and with task and motion planning approaches, resulting in a significant improvement in terms of planning time, success rate and quality of the solution path.

show abstract

Section: B Benchmarkingmentioning

confidence: 99%

“…Obj. represents the number of objects used, TP and TPU represent the approaches presented in [2] and [3], respectively. pKP and pKPU represent the p-KPIECE with and without uncertainty, respectively.…”

Section: B Benchmarkingmentioning

confidence: 99%

Randomized Physics-Based Motion Planning for Grasping in Cluttered and Uncertain Environments

Muhayyuddin

Moll

Kavraki

et al. 2018

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

show abstract

“…Some popular approaches for generating policies in POMDPs are online planning [5], [6], [7] and finding a policy offline with a point-based solver [8], [9]. Our work will use the more efficient but more approximate determinizeand-replan approach, which optimistically plans in a determinized version of the environment, brought about by (for instance) assuming that the maximum likelihood observation is always obtained [10], [11]. The agent executes this plan and replans any time it receives an observation contradicting the optimistic assumptions made.…”

Section: A Pomdps and Belief Statesmentioning

confidence: 99%

Integrating Human-Provided Information into Belief State Representation Using Dynamic Factorization

Chitnis¹,

Kaelbling²,

Lozano-Pérez³

2018

2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

Self Cite

View full text Add to dashboard Cite

In partially observed environments, it can be useful for a human to provide the robot with declarative information that represents probabilistic relational constraints on properties of objects in the world, augmenting the robot's sensory observations. For instance, a robot tasked with a searchand-rescue mission may be informed by the human that two victims are probably in the same room. An important question arises: how should we represent the robot's internal knowledge so that this information is correctly processed and combined with raw sensory information? In this paper, we provide an efficient belief state representation that dynamically selects an appropriate factoring, combining aspects of the belief when they are correlated through information and separating them when they are not. This strategy works in open domains, in which the set of possible objects is not known in advance, and provides significant improvements in inference time over a static factoring, leading to more efficient planning for complex partially observed tasks. We validate our approach experimentally in two open-domain planning problems: a 2D discrete gridworld task and a 3D continuous cooking task. A supplementary video can be found at http://tinyurl.com/chitnis-iros-18.

show abstract

“…When only observations are uncertain and actions are deterministic, Hadfield-Menell, Groshev, Chitnis, and Abbeel (2015) can apply maximum-likelihood estimation to transform problems to purely deterministic representations, which can then be solved by off-the-shelf planners.…”

Section: Planning In Belief Spacementioning

confidence: 99%

KABouM: Knowledge-Level Action and Bounding Geometry Motion Planner

Gaschler¹,

Petrick²,

Khatib³

et al. 2018

jair

View full text Add to dashboard Cite

For robots to solve real world tasks, they often require the ability to reason about both symbolic and geometric knowledge. We present a framework, called KABouM, for integrating knowledge-level task planning and motion planning in a bounding geometry. By representing symbolic information at the knowledge level, we can model incomplete information, sensing actions and information gain; by representing all geometric entitiesobjects, robots and swept volumes of motions-by sets of convex polyhedra, we can efficiently plan manipulation actions and raise reasoning about geometric predicates, such as collisions, to the symbolic level. At the geometric level, we take advantage of our bounded convex decomposition and swept volume computation with quadratic convergence, and fast collision detection of convex bodies. We evaluate our approach on a wide set of problems using real robots, including tasks with multiple manipulators, sensing and branched plans, and mobile manipulation.

show abstract

Modular task and motion planning in belief space

Cited by 45 publications

References 12 publications

Randomized Physics-Based Motion Planning for Grasping in Cluttered and Uncertain Environments

Randomized Physics-Based Motion Planning for Grasping in Cluttered and Uncertain Environments

Integrating Human-Provided Information into Belief State Representation Using Dynamic Factorization

KABouM: Knowledge-Level Action and Bounding Geometry Motion Planner

Contact Info

Product

Resources

About