Academia.eduAcademia.edu

Outline

Title
Abstract
Introduction
Related Work
Background
Experimental Method
Results and Discussion
Conclusions
References
All Topics
Computer Science
Artificial Intelligence

Grasp Learning by Sampling from Demonstration

Profile image of Justus PiaterJustus Piater

2016, ArXiv

Abstract

Robotic grasping traditionally relies on object features or shape information for learning new or applying already learned grasps. We argue however that such a strong reliance on object geometric information renders grasping and grasp learning a difficult task in the event of cluttered environments with high uncertainty where reasonable object models are not available. This being so, in this paper we thus investigate the application of model-free stochastic optimization for grasp learning. For this, our proposed learning method requires just a handful of user-demonstrated grasps and an initial prior by a rough sketch of an object's grasp affordance density, yet no object geometric knowledge except for its pose. Our experiments show promising applicability of our proposed learning method.

Related papers

DemoGrasp: Few-Shot Learning for Robotic Grasping with Human Demonstration
Lorenzo Garattoni

2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2021

The ability to successfully grasp objects is crucial in robotics, as it enables several interactive downstream applications. To this end, most approaches either compute the full 6D pose for the object of interest or learn to predict a set of grasping points. While the former approaches do not scale well to multiple object instances or classes yet, the latter require large annotated datasets and are hampered by their poor generalization capabilities to new geometries. To overcome these shortcomings, we propose to teach a robot how to grasp an object with a simple and short human demonstration. Hence, our approach neither requires many annotated images nor is it restricted to a specific geometry. We first present a small sequence of RGB-D images displaying a human-object interaction. This sequence is then leveraged to build associated hand and object meshes that represent the depicted interaction. Subsequently, we complete missing parts of the reconstructed object shape and estimate the relative transformation between the reconstruction and the visible object in the scene. Finally, we transfer the a-priori knowledge from the relative pose between object and human hand with the estimate of the current object pose in the scene into necessary grasping instructions for the robot. Exhaustive evaluations with Toyota's Human Support Robot (HSR) in real and synthetic environments demonstrate the applicability of our proposed methodology and its advantage in comparison to previous approaches.

View PDFchevron_right
Leveraging Contact Forces for Learning to Grasp
miroslav bogdanovic

2019 International Conference on Robotics and Automation (ICRA), 2019

Grasping objects under uncertainty remains an open problem in robotics research. This uncertainty is often due to noisy or partial observations of the object pose or shape. To enable a robot to react appropriately to unforeseen effects, it is crucial that it continuously takes sensor feedback into account. While visual feedback is important for inferring a grasp pose and reaching for an object, contact feedback offers valuable information during manipulation and grasp acquisition. In this paper, we use model-free deep reinforcement learning to synthesize control policies that exploit contact sensing to generate robust grasping under uncertainty. We demonstrate our approach on a multi-fingered hand that exhibits more complex finger coordination than the commonly used twofingered grippers. We conduct extensive experiments in order to assess the performance of the learned policies, with and without contact sensing. While it is possible to learn grasping policies without contact sensing, our results suggest that contact feedback allows for a significant improvement of grasping robustness under object pose uncertainty and for objects with a complex shape.

View PDFchevron_right
Domain Randomization and Generative Models for Robotic Grasping
Lukas Biewald

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

View PDFchevron_right
Learning continuous grasp affordances by sensorimotor exploration
Mila Popovic

From Motor Learning to …, 2010

We develop means of learning and representing object grasp affordances probabilistically. By grasp affordance, we refer to an entity that is able to assess whether a given relative object-gripper configuration will yield a stable grasp. These affordances are represented with grasp densities, continuous probability density functions defined on the space of 3D positions and orientations. Grasp densities are registered with a visual model of the object they characterize. They are exploited by aligning them to a target object using visual pose estimation. Grasp densities are refined through experience: A robot "plays" with an object by executing grasps drawn randomly for the object's grasp density. The robot then uses the outcomes of these grasps to build a richer density through an importance sampling mechanism. Initial grasp densities, called hypothesis densities, are bootstrapped from grasps collected using a motion capture system, or from grasps generated from the visual model of the object. Refined densities, called empirical densities, represent affordances that have been confirmed through physical experience. The applicability of our method is demonstrated by producing empirical densities for two object with a real robot and its 3-finger hand. Hypothesis densities are created from visual cues and human demonstration.

View PDFchevron_right
Grasp Generalization Via Predictive Parts
Justus Piater

2011

View PDFchevron_right
Active and Transfer Learning of Grasps by Sampling from Demonstration
Justus Piater

ArXiv, 2016

We guess humans start acquiring grasping skills as early as at the infant stage by virtue of two key processes. First, infants attempt to learn grasps for known objects by imitating humans. Secondly, knowledge acquired during this process is reused in learning to grasp novel objects. We argue that these processes of active and transfer learning boil down to a random search of grasps on an object, suitably biased by prior experience. In this paper we introduce active learning of grasps for known objects as well as transfer learning of grasps for novel objects grounded on kernel adaptive, mode-hopping Markov Chain Monte Carlo. Our experiments show promising applicability of our proposed learning methods.

View PDFchevron_right
Learning to grasp under uncertainty
Stefan Schaal

2011

Abstract We present an approach that enables robots to learn motion primitives that are robust towards state estimation uncertainties. During reaching and preshaping, the robot learns to use line manipulation strategies to maneuver the object into a pose at which closing the hand to perform the grasp is more likely to succeed.

View PDFchevron_right
Robot Grasp Learning by Demonstration without Predefined Rules
Rafa Puerto

International Journal of Advanced Robotic Systems, 2011

View PDFchevron_right
Data-Driven Grasp Synthesis—A Survey
Tamim Asfour

IEEE Transactions on Robotics, 2000

We review the work on data-driven grasp synthesis and the methodologies for sampling and ranking candidate grasps. We divide the approaches into three groups based on whether they synthesize grasps for known, familiar or unknown objects. This structure allows us to identify common object representations and perceptual processes that facilitate the employed data-driven grasp synthesis technique. In the case of known objects, we concentrate on the approaches that are based on object recognition and pose estimation. In the case of familiar objects, the techniques use some form of a similarity matching to a set of previously encountered objects. Finally, for the approaches dealing with unknown objects, the core part is the extraction of specific features that are indicative of good grasps. Our survey provides an overview of the different methodologies and discusses open problems in the area of robot grasping. We also draw a parallel to the classical approaches that rely on analytic formulations.

View PDFchevron_right
Learning to grasp objects with multiple contact points
Arda Kara

2010 IEEE International Conference on Robotics and Automation, 2010

We consider the problem of grasping novel objects and its application to cleaning a desk. A recent successful approach applies machine learning to learn one grasp point in an image and a point cloud. Although those methods are able to generalize to novel objects, they yield suboptimal results because they rely on motion planner for finger placements. In this paper, we extend their method to accommodate grasps with multiple contacts. This approach works well for many humanmade objects because it models the way we grasp objects. To further improve the grasping, we also use a method that learns the ranking between candidates. The experiments show that our method is highly effective compared to a state-of-the-art competitor.

View PDFchevron_right

References (41)

  1. J. Bohg, A. Morales, T. Asfour, and D. Kragic, "Data-driven Grasp Synthesis -A Survey," Robotics, IEEE Transactions on, vol. 30, no. 2, pp. 289-309, 2014.
  2. A. Sahbani, S. El-Khoury, and P. Bidaud, "An Overview of 3D Object Grasp Synthesis Algorithms," Robotics and Autonomous Systems, vol. 60, no. 3, pp. 326-336, 2012.
  3. R. Balasubramanian, L. Xu, P. D. Brook, J. R. Smith, and Y. Matsuoka, "Physical Human Interactive Guidance: Identifying Grasping Princi- ples from Human-planned Grasps," IEEE Transactions on Robotics, vol. 28, no. 4, pp. 899-910, 2012.
  4. W. K. Hastings, "Monte Carlo Sampling Methods Using Markov Chains and Their Applications," Biometrika, vol. 57, no. 1, pp. 97- 109, 1970.
  5. D. Sejdinovic, H. Strathmann, M. L. Garcia, C. Andrieu, and A. Gret- ton, "Kernel Adaptive Metropolis-Hastings," in Proceedings of the 31st International Conference on Machine Learning, E. P. Xing and T. Jebara, Eds., vol. 32, 2014, pp. 1665-1673.
  6. C. Sminchisescu and M. Welling, "Generalized darting Monte Carlo," Pattern Recognition, vol. 44, no. 10-11, pp. 2738-2748, 2011.
  7. S. Ekvall and D. Kragić, "Interactive Grasp Learning Based on Human Demonstration," in IEEE International Conference on Robotics and Automation, ICRA'04, vol. 4. IEEE, 2004, pp. 3519-3524.
  8. --, "Grasp Recognition for Programming by Demonstration," in IEEE International Conference on Robotics and Automation, ICRA'05. IEEE, 2005, pp. 748-753.
  9. Fig.
  10. Dispersion plots for all runs w.r.t. c relative to the number of successfully sampled grasps. The cirlces' sizes denote the surface area of the convex hull of the set of successfully sampled grasps, thence describing the exploratory behavior of the sampler for a given value of c. Orange denotes runs with impartial, blue with weak, and green with strong bias. Due to space restrictions values for c were modulated with increments of 0.01. Best viewed in color.
  11. H. Kjellström, J. Romero, and D. Kragić, "Visual Recognition of Grasps for Human-to-Robot Mapping," in IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS'08. IEEE, 2008, pp. 3192-3199.
  12. J. Romero, H. Kjellström, and D. Kragić, "Human-to-Robot Mapping of Grasps," in IEEE/RSJ International Conference on Intelligent Robots and Systems, Workshop on Grasp and Task Learning by Imitation, IROS'08, 2008.
  13. --, "Modeling and Evaluation of Human-to-Robot Mapping of Grasps," in International Conference on Advanced Robotics, ICAR'09. IEEE, 2009, pp. 1-6.
  14. J. Aleotti and S. Caselli, "Grasp Recognition in Virtual Reality for Robot Pregrasp Planning by Demonstration," in IEEE International Conference on Robotics and Automation, ICRA'06. IEEE, 2006, pp. 2801-2806.
  15. --, "Robot Grasp Synthesis from Virtual Demonstration and Topology-preserving Environment Reconstruction," in IEEE/RSJ In- ternational Conference on Intelligent Robots and Systems, IROS'07. IEEE, 2007, pp. 2692-2697.
  16. Y. Lin and Y. Sun, "Robot Grasp Planning Based on Demonstrated Grasp Strategies," The International Journal of Robotics Research, pp. 26-42, 2014.
  17. R. Zöllner, O. Rogalla, R. Dillmann, and J. Zoellner, "Dynamic Grasp Recognition Within the Framework of Programming by Demon- stration," in IEEE International Workshop on Robot and Human Interactive Communication. IEEE, 2001, pp. 418-423.
  18. Y. Li and N. S. Pollard, "A Shape Matching Algorithm for Syn- thesizing Humanlike Enveloping Grasps," in IEEE-RAS International Conference on Humanoid Robots, HUMANOIDS'05. IEEE, 2005, pp. 442-449.
  19. F. Kyota, T. Watabe, S. Saito, and M. Nakajima, "Detection and Evalu- ation of Grasping Positions for Autonomous Agents," in International Conference on Cyberworlds. IEEE, 2005, pp. 460-468.
  20. A. Herzog, P. Pastor, M. Kalakrishnan, L. Righetti, J. Bohg, T. As- four, and S. Schaal, "Learning of Grasp Selection Based on Shape- templates," Autonomous Robots, vol. 36, no. 1-2, pp. 51-65, 2014.
  21. S. Ekvall and D. Kragić, "Learning and Evaluation of the Approach Vector for Automatic Grasp Generation and Planning," in IEEE Inter- national Conference on Robotics and Automation, ICRA'07. IEEE, 2007, pp. 4715-4720.
  22. J. Tegin, S. Ekvall, D. Kragic, J. Wikander, and B. Iliev, "Demonstration-based Learning and Control for Automatic Grasping," Intelligent Service Robotics, vol. 2, no. 1, pp. 23-30, 2009.
  23. J. Aleotti and S. Caselli, "Part-based Robot Grasp Planning from Human Demonstration," in IEEE International Conference on Robotics and Automation, ICRA'11. IEEE, 2011, pp. 4554-4560.
  24. K. Hsiao and T. Lozano-Perez, "Imitation Learning of Whole-body Grasps," in IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS'06. IEEE, 2006, pp. 5657-5662.
  25. M. Do, T. Asfour, and R. Dillmann, "Towards a Unifying Grasp Representation for Imitation Learning on Humanoid Robots," in IEEE International Conference on Robotics and Automation, ICRA'11. IEEE, 2011, pp. 482-488.
  26. O. Kroemer, R. Detry, J. Piater, and J. Peters, "Combining Active Learning and Reactive Control for Robot Grasping," Robotics and Autonomous Systems, vol. 58, no. 9, pp. 1105-1116, 2010.
  27. P. Pastor, L. Righetti, M. Kalakrishnan, and S. Schaal, "Online Move- ment Adaptation based on Previous Sensor experiences," in IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS'11. IEEE, 2011, pp. 365-371.
  28. E. Oztop and M. A. Arbib, "Schema Design and Implementation of the Grasp-related Mirror Neuron System," Biological Cybernetics, vol. 87, no. 2, pp. 116-140, 2002.
  29. M. Hueser, T. Baier, and J. Zhang, "Learning of Demonstrated Grasp- ing Skills by Stereoscopic Tracking of Human Head Configuration," in IEEE International Conference on Robotics and Automation, ICRA'06. IEEE, 2006, pp. 2795-2800.
  30. C. de Granville, J. Southerland, and A. H. Fagg, "Learning Grasp Affordances Through Human Demonstration," in International Con- ference on Development and Learning, ICDL'06. IEEE, 2006.
  31. D. R. Faria, R. Martins, J. Lobo, and J. Dias, "Extracting Data from Human Manipulation of Objects Towards Improving Autonomous Robotic Grasping," Robotics and Autonomous Systems, vol. 60, no. 3, pp. 396-410, 2012.
  32. R. Detry, D. Kraft, O. Kroemer, L. Bodenhagen, J. Peters, N. Krüger, and J. Piater, "Learning Grasp Affordance Densities," Paladyn, vol. 2, no. 1, pp. 1-17, 2011.
  33. J. D. Sweeney and R. Grupen, "A Model of Shared Grasp Affordances From Demonstration," in IEEE-RAS International Conference on Humanoids Robots, HUMANOIDS'07. IEEE, 2007, pp. 27-35.
  34. M. Kopicki, R. Detry, F. Schmidt, C. Borst, R. Stolkin, and J. Wyatt, "Learning Dexterous Grasps that Generalise to Novel Objects by Com- bining Hand and Contact Models," in IEEE International Conference on Robotics and Automation, ICRA'14., May 2014, pp. 5358-5365.
  35. M. Kopicki, R. Detry, M. Adjigble, R. Stolkin, A. Leonardis, and J. L. Wyatt, "One-shot Learning and Generation of Dexterous Grasps for Novel Objects," vol. 35, no. 8, pp. 959-976, 2016.
  36. A. T. Miller and P. K. Allen, "Examples of 3D Grasp Quality Computations," in IEEE International Conference on Robotics and Automation, 1999. Proceedings, ICRA'99., vol. 2. IEEE, 1999, pp. 1240-1246.
  37. P. Zech, H. Xiong, and J. Piater, "Rotation Optimization on the Unit Quaternion Manifold and its Application for Robotic Grasping," in IMA Conference on Mathematics of Robotics. IMA, 2015, to appear.
  38. M. Lang, M. Kleinsteuber, O. Dunkley, and S. Hirche, "Gaussian Process Dynamical Models Over Dual Quaternions," in European Control Conference (ECC), 2015. IEEE, 2015, pp. 2847-2852.
  39. L.-P. Ellekilde and J. A. Jorgensen, "RobWork: A Flexible Toolbox for Robotics Research and Education," in 41st International Symposium on Robotics (ISR) and 2010 6th German Conference on Robotics (ROBOTIK). VDE, 2010, pp. 1-7.
  40. C. Andrieu and J. Thoms, "A Ttutorial on Adaptive MCMC," Statistics and Computing, vol. 18, no. 4, pp. 343-373, 2008.
  41. A. Roby-Brami, N. Bennis, M. Mokhtari, and P. Baraduc, "Hand Orientation for Grasping Depends on the Direction of the Reaching Movement," Brain Research, vol. 869, no. 1-2, pp. 121 -129, 2000.

Related papers

Refining grasp affordance models by experience
Renaud Detry

2010

We present a method for learning object grasp affordance models in 3D from experience, and demonstrate its applicability through extensive testing and evaluation on a realistic and largely autonomous platform. Grasp affordance refers here to relative object-gripper configurations that yield stable grasps. These affordances are represented probabilistically with grasp densities, which correspond to continuous density functions defined on the space of 6D gripper poses. A grasp density characterizes an object's grasp affordance; densities are linked to visual stimuli through registration with a visual model of the object they characterize. We explore a batch-oriented, experience-based learning paradigm where grasps sampled randomly from a density are performed, and an importance-sampling algorithm learns a refined density from the outcomes of these experiences. The first such learning cycle is bootstrapped with a grasp density formed from visual cues. We show that the robot effectively applies its experience by downweighting poor grasp solutions, which results in increased success rates at subsequent learning cycles. We also present success rates in a practical scenario where a robot needs to repeatedly grasp an object lying in an arbitrary pose, where each pose imposes a specific reaching constraint, and thus forces the robot to make use of the entire grasp density to select the most promising achievable grasp.

View PDFchevron_right
Learning grasp affordance densities
Renaud Detry

Paladyn, 2011

We address the issue of learning and representing object grasp affordance models. We model grasp affordances with continuous probability density functions (grasp densities) which link object-relative grasp poses to their success probability. The underlying function representation is nonparametric and relies on kernel density estimation to provide a continuous model. Grasp densities are learned and refined from exploration, by letting a robot “play” with an object in a sequence of grasp-and-drop actions: the robot uses visual cues to generate a set of grasp hypotheses, which it then executes and records their outcomes. When a satisfactory amount of grasp data is available, an importance-sampling algorithm turns it into a grasp density. We evaluate our method in a largely autonomous learning experiment, run on three objects with distinct shapes. The experiment shows how learning increases success rates. It also measures the success rate of grasps chosen to maximize the probability of success, given reaching constraints.

View PDFchevron_right
Learn to Grasp with Less Supervision: A Data-Efficient Maximum Likelihood Grasp Sampling Loss
Yefan Zhou

ArXiv, 2021

Robotic grasping for a diverse set of objects is essential in many robot manipulation tasks. One promising approach is to learn deep grasping models from training datasets of object images and grasp labels. Approaches in this category require millions of data to train deep models. However, empirical grasping datasets typically consist of sparsely labeled samples (i.e., a limited number of successful grasp labels in each image). This paper proposes a Maximum Likelihood Grasp Sampling Loss (MLGSL) to tackle the data sparsity issue. The proposed method supposes that successful grasp labels are sampled from a ground-truth grasp distribution and aims to recover the ground-truth map. MLGSL is utilized for training a fully convolutional network that detects thousands of grasps simultaneously. Training results suggest that models based on MLGSL can learn to grasp with datasets composing of 2 labels per image, which implies that it is 8× more data-efficient than current state-of-the-art tech...

View PDFchevron_right
Accelerating Grasp Exploration by Leveraging Learned Priors
Ashwin Balakrishna

2020 IEEE 16th International Conference on Automation Science and Engineering (CASE)

View PDFchevron_right
Learning object-specific grasp affordance densities
Mila Popovic

… and Learning, 2009. …, 2009

This paper addresses the issue of learning and representing object grasp affordances, i.e. object-gripper relative configurations that lead to successful grasps. The purpose of grasp affordances is to organize and store the whole knowledge that an agent has about the grasping of an object, in order to facilitate reasoning on grasping solutions and their achievability. The affordance representation consists in a continuous probability density function defined on the 6D gripper pose space -3D position and orientation -, within an object-relative reference frame. Grasp affordances are initially learned from various sources, e.g. from imitation or from visual cues, leading to grasp hypothesis densities. Grasp densities are attached to a learned 3D visual object model, and pose estimation of the visual model allows a robotic agent to execute samples from a grasp hypothesis density under various object poses. Grasp outcomes are used to learn grasp empirical densities, i.e. grasps that have been confirmed through experience. We show the result of learning grasp hypothesis densities from both imitation and visual cues, and present grasp empirical densities learned from physical experience by a robot.

View PDFchevron_right

Related topics

  • Computer Science
  • Artificial Intelligence
    • 580 California St., Suite 400
    San Francisco, CA, 94104
    © 2025 Academia. All rights reserved