25,346 Hits in 5.7 sec

Model-free and learning-free grasping by Local Contact Moment matching

Maxime Adjigble, Naresh Marturi, Valerio Ortenzi, Vijaykumar Rajasekaran, Peter Corke, Rustam Stolkin
2018 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)  
This metric is model-free, and does not need to be learned from training data. • Exploitation of the kinematics of the robot to select a subset of the graspable points, first identified by LoCoMo, that  ...  ., model-free). Using LoCoMo as a fitness function, a point-cloud surface can be efficiently searched for good matches to finger surface geometry.  ... 
doi:10.1109/iros.2018.8594226 dblp:conf/iros/AdjigbleMORCS18 fatcat:24hbns357rdspd4hbybgsvaxh4


2005 International Journal of Humanoid Robotics  
Individual robots must exploit declarative structure for planning and must learn procedural strategies that work in recognizable contexts.  ...  We present several pieces of an overall framework in which a robot learns situated policies for control that exploit existing control knowledge and extend its scope.  ...  The work described herein was supported in part by the NSF (CDA 9703217), DARPA-MARS (DABT63-99-1-0004), NASA (NAG9-1445#1), and the University of Massachusetts.  ... 
doi:10.1142/s0219843605000491 fatcat:quucpz6uazfznfcgrr6ywkjltm

InsertionNet – A Scalable Solution for Insertion [article]

Oren Spector, Dotan Di Castro
2021 arXiv   pre-print
Complicated assembly processes can be described as a sequence of two main activities: grasping and insertion.  ...  However, these approaches might be problematic in contact-rich tasks since interaction might endanger the robot and its equipment.  ...  ACKNOWLEDGMENT We thank the anonymous reviewers and Mrs. Dana Rip for their very helpful comments that improved this manuscript.  ... 
arXiv:2104.14223v1 fatcat:qfk65l2h4nhqfhhbjupz2y5sku

SpectGRASP: Robotic Grasping by Spectral Correlation [article]

Maxime Adjigble, Cristiana de Farias, Rustam Stolkin, Naresh Marturi
2021 arXiv   pre-print
We then use our previous work, Local Contact Moment (LoCoMo) similarity metric, to sequentially rank the generated grasps such that the one with maximum likelihood is executed.  ...  Given a point cloud of an object, SpectGRASP extracts contact points on the object's surface matching the hand configuration. It neither requires offline training nor a-priori object models.  ...  All authors are with the Extreme Robotics Laboratory, School of Metallurgy and Materials, University of Birmingham, Edgbaston, B15 2TT, UK.  ... 
arXiv:2107.12492v1 fatcat:jkhjrdo3rfeo5gvaurcmd3y34i

Contact localization on grasped objects using tactile sensing

Artem Molchanov, Oliver Kroemer, Zhe Su, Gaurav S. Sukhatme
2016 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)  
In this paper, we present a data-driven approach for detecting and localizing contacts between a grasped object and the environment using tactile sensing.  ...  We also compare the neural networks with Gaussian process regression and support vector machine classification with spatiotemporal hierarchical matching pursuit feature learning.  ...  BioTac sensors used in this research were kindly provided by SynTouch LLC. This work was funded in part by the NSF under grant CNS-1213128 and the ONR under grant N00014-14-1-0536.  ... 
doi:10.1109/iros.2016.7759058 dblp:conf/iros/MolchanovKSS16 fatcat:rjuzgaqg3zc3rbk7ewxnoe644e

Autonomous soft hand grasping – Literature review [article]

Tai Hoang
2022 arXiv   pre-print
We start by briefly discussing the analytic methods that mainly exploit the hand dynamic information. Then, data-driven approaches will be our main focus.  ...  Autonomous grasping remains challenging as unlike humans, robots do not possess a sophisticated sensing nor delicate interaction capability with the real environment.  ...  In this section, we will discuss two popular directions, discriminative approaches and model-free grasp synthesis.  ... 
arXiv:2203.04762v1 fatcat:7z22gtloyzaa5mtrz47a4w2b3u

GIFT: Generalizable Interaction-aware Functional Tool Affordances without Labels [article]

Dylan Turpin, Liquan Wang, Stavros Tsogkas, Sven Dickinson, Animesh Garg
2021 arXiv   pre-print
GIFT outperforms baselines on all tasks and matches a human oracle on two of three tasks using novel tools.  ...  In our experiments, we show that GIFT can leverage a sparse keypoint representation to predict grasp and interaction points to accommodate multiple tasks, such as hooking, reaching, and hammering.  ...  AG was supported, in part, by the CIFAR AI Chair Grant. The authors would also like to acknowledge the feedback from members of the PAIR research group at UofT.  ... 
arXiv:2106.14973v1 fatcat:g4jzc6pnzrcclmfzwpimtuja5a

Learning Object Affordances: From Sensory--Motor Coordination to Imitation

Luis Montesano, Manuel Lopes, Alexandre Bernardino, JosÉ Santos-Victor
2008 IEEE Transactions on robotics  
Since learning is based on a probabilistic model, the approach is able to deal with uncertainty, redundancy, and irrelevant information.  ...  We demonstrate successful learning in the real world by having an humanoid robot interacting with objects. We illustrate the benefits of the acquired knowledge in imitation games.  ...  (b) CPD of height given the robot obtained a long contact (successful grasp). Fig. 8 . 8 This figure shows the affordance model estimated by the K2 algorithm and the MCMC.  ... 
doi:10.1109/tro.2007.914848 fatcat:6r3eojjwqretlmy4a4ydsymmge

Addressing perception uncertainty induced failure modes in robotic bin-picking

Krishnanand N. Kaipa, Akshaya S. Kankanhalli-Nagendra, Nithyananda B. Kumbla, Shaurya Shriyam, Srudeep Somnaath Thevendria-Karthic, Jeremy A. Marvel, Satyandra K. Gupta
2016 Robotics and Computer-Integrated Manufacturing  
Our approach estimates the confidence in the part match provided by an automated perception system, which is used to detect perception failures.  ...  We identify the main failure modes at various stages of the bin-picking task and present methods to recover from them.  ...  This work is supported in part by National Science Foundation Grants #1200087 and #1527220 and National Institute of Standards and Technology Cooperative Agreement #70NANB15H250.  ... 
doi:10.1016/j.rcim.2016.05.002 fatcat:wommeugndfd6za5zk5ie7rzav4

Neural Correlates of Internal-Model Loading

Lulu L.C.D. Bursztyn, G. Ganesh, Hiroshi Imamizu, Mitsuo Kawato, J. Randall Flanagan
2006 Current Biology  
By comparing the MOVE task with the FREE and ISO tasks, we thought to isolate loading of the internal model of the object dynamics.  ...  Recent behavioral studies have shown that once learned, an internal model of an object with novel dynamics can be rapidly recruited and derecruited as the object is grasped and released [10] [11] [12]  ...  The procedure was approved by a local ethics committee. All subjects performed three different trial conditions: MOVE, ISO, and FREE.  ... 
doi:10.1016/j.cub.2006.10.051 pmid:17174919 fatcat:2da2oiog3zcwrns2sqht6opjpu

Dexterous grasping under shape uncertainty

Miao Li, Kaiyu Hang, Danica Kragic, Aude Billard
2016 Robotics and Autonomous Systems  
A compliant nger closing scheme is devised by exploiting both the object shape uncertainty and tactile sensing at ngertips.  ...  An important challenge in robotics is to achieve robust performance in object grasping and manipulation, dealing with noise and uncertainty.  ...  At this moment, to improve the performance, we adopt a local derivative-free optimization technique, called Constrained Optimization by Linear Approximation (COBYLA) [60] .  ... 
doi:10.1016/j.robot.2015.09.008 fatcat:c5hmrb2qgzfhdlkpmbmhbqdbx4

Dynamic grasp and trajectory planning for moving objects

Naresh Marturi, Marek Kopicki, Alireza Rastegarpanah, Vijaykumar Rajasekaran, Maxime Adjigble, Rustam Stolkin, Aleš Leonardis, Yasemin Bekiroglu
2018 Autonomous Robots  
While the object is being arbitrarily moved by the human co-worker, a set of likely grasps, generated by a learned grasp planner, are evaluated online to generate a feasible grasp with respect to both:  ...  the current configuration of the robot respecting the target grasp; and the constraints of finding a collision-free trajectory to reach that configuration.  ...  , and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.  ... 
doi:10.1007/s10514-018-9799-1 fatcat:zhxisgmkknfpjbqeo2tfc7a5sy

Model-based contextual policy search for data-efficient generalization of robot skills

Andras Kupcsik, Marc Peter Deisenroth, Jan Peters, Ai Poh Loh, Prahlad Vadakkepat, Gerhard Neumann
2017 Artificial Intelligence  
However, the majority of current contextual policy search approaches are model-free and require a high number of interactions with the robot and its environment.  ...  Our approach is based on learned probabilistic forward models and information theoretic policy search.  ...  Marc Peter Deisenroth was supported by an Imperial College Junior Research Fellowship.  ... 
doi:10.1016/j.artint.2014.11.005 fatcat:faefrs36vbgurknztnotyajlhi

Planning for Dexterous Ungrasping: Secure Ungrasping through Dexterous Manipulation [article]

Chung Hee Kim, Ka Hei Mak, Jungwon Seo
2021 arXiv   pre-print
The game of Go offers an example: consider how the player would typically place an initially pinch-grasped stone onto the board through the dexterous interaction between the fingers, the stone, and the  ...  It refers to the capability of securely transferring a grasped object from the gripper to the robot's environment, i.e. the inverse of grasping or picking, through dexterous manipulation.  ...  Second, we define the free configuration space C free , i.e. the search space of the algorithm, as: C free = (C ∩ C grasp ) \ C obs (3) where C obs is the obstacle space and C grasp is the collection of  ... 
arXiv:2108.13580v1 fatcat:yot723vptzd45fbxu4th6q3iaq

Grasp Planning Using Low Dimensional Subspaces [chapter]

Peter K. Allen, Matei Ciocarlie, Corey Goldfeder
2014 Springer Tracts in Advanced Robotics  
ACKNOWLEDGMENT The authors would like to thank Hao Dang for his help in building the Columbia Grasp Database.  ...  [40] generated 2D renderings of a large set of example objects, and learned a model-free mapping from images to graspable features.  ...  However, other local contact models can also be used.  ... 
doi:10.1007/978-3-319-03017-3_24 fatcat:edo4tv4gjjer3f5cluqylr7x5q
« Previous Showing results 1 — 15 out of 25,346 results