Perception for mobile manipulation and grasping using active stereo

Rusu, Radu Bogdan; Holzbach, Andreas; Diankov, Rosen; Beetz, Michael

doi:10.1109/ichr.2009.5379597

Cited by 43 publications

(32 citation statements)

References 16 publications

(22 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This problem can be mitigated by computing independent contact regions (ICRs) (Ponce & Faverjon, 1995), i.e., maximal segments of the object's surface where the fingers can be applied while maintaining force closure. Yet, ICRs still suffer from other shortcomings of force analysis, such as difficulties in estimating shape, mass or friction parameters (Rusu et al, 2009). …”

Section: Ijrr -(-)mentioning

confidence: 99%

One-shot learning and generation of dexterous grasps for novel objects

Kopicki

Detry

Adjigble

et al. 2015

The International Journal of Robotics Research

115

163

View full text Add to dashboard Cite

This paper presents a method for one-shot learning of dexterous grasps, and grasp generation for novel objects. A model of each grasp type is learned from a single kinesthetic demonstration, and several types are taught. These models are used to select and generate grasps for unfamiliar objects. Both the learning and generation stages use an incomplete point cloud from a depth camera -no prior model of object shape is used. The learned model is a product of experts, in which experts are of two types. The first is a contact model and is a density over the pose of a single hand link relative to the local object surface. The second is the hand configuration model and is a density over the whole hand configuration. Grasp generation for an unfamiliar object optimises the product of these two model types, generating thousands of grasp candidates in under 30 seconds. The method is robust to incomplete data at both training and testing stages. When several grasp types are considered the method selects the highest likelihood grasp across all the types. In an experiment, the training set consisted of five different grasps, and the test set of forty-five previously unseen objects. The success rate of the first choice grasp is 84.4% or 77.7% if seven views or a single view of the test object are taken, respectively. Keywords learning, dexterous grasping * Authors Kopicki, Detry and Wyatt are identified as the primary authors of this work. Kopicki is identified as the first author. † Corresponding author; e-mail: jlw@cs.bham.ac.uk 2 IJRR -(-) Fig. 1. Leftmost image: Objects used, the four objects on the left were used solely for training, the remaining forty three objects on the right were solely used as novel test objects. Rightmost image: The Boris manipulation platform on which the experiments reported were carried out.

show abstract

Section: Ijrr -(-)mentioning

confidence: 99%

One-shot learning and generation of dexterous grasps for novel objects

Kopicki

Detry

Adjigble

et al. 2015

The International Journal of Robotics Research

115

163

View full text Add to dashboard Cite

show abstract

“…With access to depth sensors becoming easier everyday, increasingly many methods rely on depth as input for various applications, such as autonomous driving [17], augmented reality [19] and personal robotics [20]. Unfortunately, depth sensors are not perfect; they typically produce relatively sparse measurements with large holes.…”

Section: Related Workmentioning

confidence: 99%

“…Fixing the semantic variable s h , and writing the objective in Eq. (20) in the standard form yields…”

Section: Optimization Wrt the Hidden Layer S H U Hmentioning

confidence: 99%

See 1 more Smart Citation

Building Scene Models by Completing and Hallucinating Depth and Semantics

Liu

Salzmann

2016

Computer Vision – ECCV 2016

View full text Add to dashboard Cite

Abstract. Building 3D scene models has been a longstanding goal of computer vision. The great progress in depth sensors brings us one step closer to achieving this in a single shot. However, depth sensors still produce imperfect measurements that are sparse and contain holes. While depth completion aims at tackling this issue, it ignores the fact that some regions of the scene are occluded by the foreground objects. Building a scene model would therefore require to hallucinate the depth behind these objects. In contrast with existing methods that either rely on manual input, or focus on the indoor scenario, we introduce a fully-automatic method to jointly complete and hallucinate depth and semantics in challenging outdoor scenes. To this end, we develop a two-layer model representing both the visible information and the hidden one. At the heart of our approach lies a formulation based on the Mumford-Shah functional, for which we derive an effective optimization strategy. Our experiments evidence that our approach can accurately fill the large holes in the input depth maps, segment the different kinds of objects in the scene, and hallucinate the depth and semantics behind the foreground objects.

show abstract

“…The current trend is to incorporate sensor information for grasp planning and synthesis, such as vision [2,3,4,5,6] or range sensors [7]. In this line, several approaches have also adopted machine learning techniques to determine the relevant features that indicate a successful grasp [8,4,9,10].…”

Section: State Of the Artmentioning

confidence: 99%

Visually-guided manipulation techniques for robotic autonomous underwater panel interventions

Peñalver

Pérez

Fernández

et al. 2015

Annual Reviews in Control

View full text Add to dashboard Cite

The long term of this ongoing research has to do with increasing the autonomy levels for underwater intervention missions. Bearing in mind that the specific mission to face has been the intervention on a panel, in this paper some results in different development stages are presented by using the real mechatronics and the panel mockup. Furthermore, some details are highlighted describing two methodologies implemented for the required visually-guided manipulation algorithms, and also a roadmap explaining the different testbeds used for experimental validation, in increasing complexity order, are presented. It is worth mentioning that the aforementioned results would be impossible without previous generated know-how for both, the complete developed mechatronics for the autonomous underwater vehicle for intervention, and the required 3D simulation tool. In summary, thanks to the implemented approach, the intervention system is able to control the way in which the gripper approximates and manipulates the two panel devices (i.e. a valve and a connector) in autonomous manner and, results in different scenarios demonstrate the reliability and feasibility of this autonomous intervention system in water tank and pool conditions.

show abstract

Perception for mobile manipulation and grasping using active stereo

Cited by 43 publications

References 16 publications

One-shot learning and generation of dexterous grasps for novel objects

One-shot learning and generation of dexterous grasps for novel objects

Building Scene Models by Completing and Hallucinating Depth and Semantics

Visually-guided manipulation techniques for robotic autonomous underwater panel interventions

Contact Info

Product

Resources

About