A Data-Driven Statistical Framework for Post-Grasp Manipulation

Paolini, Robert; Rodríguez, Alberto; Srinivasa, Siddhartha S.; Mason, Matthew T.

doi:10.1007/978-3-319-00065-7_29

“…The approach presented by Jiang [14], [15] uses learning on hand-designed features and successfully places known objects stably 98% of the time and new objects 82% of the time. Paolini [16] estimates the probability of a successful placement, then attempts to solve for the most likely placement location, given a grasped object. Fu [17] use hand chosen features to try and find the upright orientation of man-made objects.…”

Section: Related Work a Robotic Placingmentioning

confidence: 99%

Learning to Place Objects Onto Flat Surfaces in Upright Orientations

Newbury

¹

,

He

²

,

Cosgun

³

et al. 2021

IEEE Robot. Autom. Lett.

View full text Add to dashboard Cite

We study the problem of placing a grasped object on an empty flat surface in a human-preferred orientation, such as placing a cup on its bottom rather than on its side. We aim to find the required object rotation such that when the gripper is opened after the object makes a contact with the surface, the object would be stably placed in the desired orientation. We use two neural networks in an iterative fashion. At every iteration, Placement Rotation CNN (PR-CNN) estimates the required object rotation which is executed by the robot, and then Placement Stability CNN (PS-CNN) estimates if the object would be stable if it is placed in its current orientation. In simulation experiments, our approach places objects in humanpreferred orientations with a success rate of 86.1% using a dataset of 18 everyday objects. A real world implementation is presented, which serves as a proof-of-concept for direct sim-toreal transfer. We observe that sometimes it is impossible to place a grasped object in a desired orientation without re-grasping, which motivates future research for grasping with intention to place objects.

show abstract

“…Gaussian Processes (GPs) are a powerful tool to model the dynamics of complex systems [5,8,9], and have been applied to different aspects of robotics including planning and control [10,11,12], system identification [8,13,14], or filtering [6,7,15]. In this work we study the problem of accurate propagation and filtering of the state of a stochastic dynamic system.…”

Section: Related Workmentioning

confidence: 99%

GP-SUM. Gaussian Process Filtering of non-Gaussian Beliefs

Bauzá

¹

,

Rodríguez

²

2020

Springer Proceedings in Advanced Robotics

Self Cite

0

View full text Add to dashboard Cite

This work studies the problem of stochastic dynamic filtering and state propagation with complex beliefs. The main contribution is GP-SUM, a filtering algorithm tailored to dynamic systems and observation models expressed as Gaussian Processes (GP), and to states represented as a weighted Sum of Gaussians. The key attribute of GP-SUM is that it does not rely on linearizations of the dynamic or observation models, or on unimodal Gaussian approximations of the belief, hence enables tracking complex state distributions. The algorithm can be seen as a combination of a sampling-based filter with a probabilistic Bayes filter. On the one hand, GP-SUM operates by sampling the state distribution and propagating each sample through the dynamic system and observation models. On the other hand, it achieves effective sampling and accurate probabilistic propagation by relying on the GP form of the system, and the sum-of-Gaussian form of the belief. We show that GP-SUM outperforms several GP-Bayes and Particle Filters on a standard benchmark. We also demonstrate its use in a pushing task, predicting with experimental accuracy the naturally occurring non-Gaussian distributions.

show abstract

“…Gaussian Processes (GPs) are a powerful tool to model the dynamics of complex systems [5,8,9], and have been applied to different aspects of robotics including planning and control [10,11,12], system identification [8,13,14], or filtering [6,7,15]. In this work we study the problem of accurate propagation and filtering of the state of a stochastic dynamic system.…”

Section: Related Workmentioning

confidence: 99%

“…distributions. Multimodality and complex belief distributions have been experimentally observed in a variety of manipulation actions such as planar pushing [2,3], ground impacts [4], and bin-picking [5].…”

Section: Introductionmentioning

confidence: 99%

GP-SUM. Gaussian Process Filtering of non-Gaussian Beliefs

Bauzá

¹

,

Rodríguez

²

2019

EasyChair Preprints

Self Cite

View full text Add to dashboard Cite

This work studies the problem of stochastic dynamic filtering and state propagation with complex beliefs. The main contribution is GP-SUM, a filtering algorithm tailored to dynamic systems and observation models expressed as Gaussian Processes (GP), and to states represented as a weighted Sum of Gaussians. The key attribute of GP-SUM is that it does not rely on linearizations of the dynamic or observation models, or on unimodal Gaussian approximations of the belief, hence enables tracking complex state distributions. The algorithm can be seen as a combination of a sampling-based filter with a probabilistic Bayes filter. On the one hand, GP-SUM operates by sampling the state distribution and propagating each sample through the dynamic system and observation models. On the other hand, it achieves effective sampling and accurate probabilistic propagation by relying on the GP form of the system, and the sum-of-Gaussian form of the belief. We show that GP-SUM outperforms several GP-Bayes and Particle Filters on a standard benchmark. We also demonstrate its use in a pushing task, predicting with experimental accuracy the naturally occurring non-Gaussian distributions.

show abstract

A Data-Driven Statistical Framework for Post-Grasp Manipulation

Cited by 16 publications

References 19 publications

Learning to Place Objects Onto Flat Surfaces in Upright Orientations

Learning to Place Objects Onto Flat Surfaces in Upright Orientations

GP-SUM. Gaussian Process Filtering of non-Gaussian Beliefs

GP-SUM. Gaussian Process Filtering of non-Gaussian Beliefs

Contact Info

Product

Resources

About