An SVM learning approach to robotic grasping

Pelossof, Raphael; Miller, Andrew T.; Allen, Peter K.; Jebara, Tony

doi:10.1109/robot.2004.1308797

Cited by 179 publications

(132 citation statements)

References 15 publications

Supporting

Mentioning

127

Contrasting

Unclassified

Order By: Relevance

“…In [27,28,29], objects are represented as basic shape primitives and then associated with predefined grasp primitives. In [30], a support vector machine (SVM) is used to learn the grasp quality manifold for a specific hand and simple object shapes. The manifold represents the mapping from grasp parameters and object shape to the grasp quality and new optimal grasps are found through interpolation on the manifold.…”

Section: Related Workmentioning

confidence: 99%

Dexterous grasping under shape uncertainty

Hang

Kragić

et al. 2016

Robotics and Autonomous Systems

View full text Add to dashboard Cite

An important challenge in robotics is to achieve robust performance in object grasping and manipulation, dealing with noise and uncertainty. This paper presents an approach for addressing the performance of dexterous grasping under shape uncertainty. In our approach, the uncertainty in object shape is parameterized and incorporated as a constraint into grasp planning. The proposed approach is used to plan feasible hand configurations for realizing planned contacts using different robotic hands. A compliant finger closing scheme is devised by exploiting both the object shape uncertainty and tactile sensing at fingertips. Experimental evaluation demonstrates that our method improves the performance of dexterous grasping under shape uncertainty.

show abstract

Section: Related Workmentioning

confidence: 99%

Dexterous grasping under shape uncertainty

Hang

Kragić

et al. 2016

Robotics and Autonomous Systems

View full text Add to dashboard Cite

show abstract

“…Grasping strategies based on the object observation analyze its properties and learn to associate them with different grasps. Some approaches associate grasp parameters or hand shapes to object geometric features in order to find good grasps in terms of stability [61,62]. Other techniques learn to identify grasping regions in an object image [63,64].…”

Section: Systems Based On the Object Observationmentioning

confidence: 99%

“…Pelossof et al [61] used support vector machines to build a regression mapping between object shape, grasp parameters and grasp quality (Fig. 9).…”

Section: Systems Based On the Object Observationmentioning

confidence: 99%

An overview of 3D object grasp synthesis algorithms

Sahbani

El-Khoury

Bidaud

2012

Robotics and Autonomous Systems

401

214

View full text Add to dashboard Cite

a b s t r a c tThis overview presents computational algorithms for generating 3D object grasps with autonomous multi-fingered robotic hands. Robotic grasping has been an active research subject for decades, and a great deal of effort has been spent on grasp synthesis algorithms. Existing papers focus on reviewing the mechanics of grasping and the finger-object contact interactions Bicchi and Kumar (2000) [12] or robot hand design and their control Al-Gallaf et al. (1993) [70]. Robot grasp synthesis algorithms have been reviewed in Shimoga (1996) [71], but since then an important progress has been made toward applying learning techniques to the grasping problem. This overview focuses on analytical as well as empirical grasp synthesis approaches.

show abstract

“…In imitation learning, some researchers use datagloves for human demonstration. The human hand configuration is then mapped to an artificial hand workspace and the joint angles [6], [8], or hand preshapes [13], [17], [26] are learnt. Some other researchers use stereoscopy to track the hand when a demonstrator is performing a grasp [10] or to match the hand shape to a database of grasp images [20].…”

Section: Related Workmentioning

confidence: 99%

“…However they focus on different problems, such as telemanipulation [8] and human hand tracking [10], rather than real time unattended grasping. Compared to those methods which concentrate on generating a list of grasps for one object [13], [17], [24], [26], our method takes one step further: we learn a model from the list and use the model to quickly generate new grasps.…”

Section: Related Workmentioning

confidence: 99%

Learning a real time grasping strategy

Huang

El-Khoury

et al. 2013

2013 IEEE International Conference on Robotics and Automation

View full text Add to dashboard Cite

Abstract-Real time planning strategy is crucial for robots working in dynamic environments. In particular, robot grasping tasks require quick reactions in many applications such as human-robot interaction. In this paper, we propose an approach for grasp learning that enables robots to plan new grasps rapidly according to the object's position and orientation. This is achieved by taking a three-step approach. In the first step, we compute a variety of stable grasps for a given object. In the second step, we propose a strategy that learns a probability distribution of grasps based on the computed grasps. In the third step, we use the model to quickly generate grasps. We have tested the statistical method on the 9 degrees of freedom hand of the iCub humanoid robot and the 4 degrees of freedom Barrett hand. The average computation time for generating one grasp is less than 10 milliseconds. The experiments were run in Matlab on a machine with 2.8GHz processor.

show abstract

An SVM learning approach to robotic grasping

Cited by 179 publications

References 15 publications

Dexterous grasping under shape uncertainty

Dexterous grasping under shape uncertainty

An overview of 3D object grasp synthesis algorithms

Learning a real time grasping strategy

Contact Info

Product

Resources

About