{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/tactilegcn-a-graph-convolutional-network-for","title":"TactileGCN: A Graph Convolutional Network for Predicting Grasp Stability with Tactile Sensors","arxiv_id":"1901.06181","date":"2019-01-18","proceeding":null,"authors":["Alberto Garcia-Garcia","Brayan Stiven Zapata-Impata","Sergio Orts-Escolano","Pablo Gil","Jose Garcia-Rodriguez"],"abstract":"Tactile sensors provide useful contact data during the interaction with an\nobject which can be used to accurately learn to determine the stability of a\ngrasp. Most of the works in the literature represented tactile readings as\nplain feature vectors or matrix-like tactile images, using them to train\nmachine learning models. In this work, we explore an alternative way of\nexploiting tactile information to predict grasp stability by leveraging\ngraph-like representations of tactile data, which preserve the actual spatial\narrangement of the sensor's taxels and their locality. In experimentation, we\ntrained a Graph Neural Network to binary classify grasps as stable or slippery\nones. To train such network and prove its predictive capabilities for the\nproblem at hand, we captured a novel dataset of approximately 5000\nthree-fingered grasps across 41 objects for training and 1000 grasps with 10\nunknown objects for testing. Our experiments prove that this novel approach can\nbe effectively used to predict grasp stability.","url_abs":"http://arxiv.org/abs/1901.06181v1","url_pdf":"http://arxiv.org/pdf/1901.06181v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"tactilegcn-a-graph-convolutional-network-for","repo_url":"https://github.com/3dperceptionlab/biotacsp-stability-set-v2","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"graph-neural-network","task_name":"Graph Neural Network"}],"methods":[{"method_slug":"graph-neural-network","method_name":"Graph Neural Network"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1901.06181","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}