[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-253015-105":3,"detail-sidebar-cat-1-en-105":81,"doc-detail-253015-en":127},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","research-on-blindsight-technology-for-object-recognition-and-attitude-determination-based-on-tactile-pressure-analysis-cnn-based-posture-determination","Research on Blindsight Technology for Object Recognition and Attitude Determination Based on Tactile Pressure Analysis - CNN-based posture determination","","Robotic grasping performance is constrained by the need for visual guidance, since environmental lighting and occlusion limit reliable object recognition during contact. This work proposes object recognition and posture determination from tactile contact-force feedback by building a contact surface deformation model grounded in Hertz contact and elastoplastic unloading theory. A vector force array dimensionality reduction method decomposes the 3D contact surface into force planes and a two-dimensional scalar array, enabling CNN learning of contact-mechanical features. Experiments validate millimeter-level grasp-position accuracy and posture-angle error under 1 degree for power-system scenario requirements, supporting non-visual humanoid end perception.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/template/","Template",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/template/presentations/","Presentations",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/template/research-on-blindsight-technology-for-object-recognition-and-attitude-determination-based-on-tactile-pressure-analysis-cnn-based-posture-determination/253015/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/research-on-blindsight-technology-for-object-recognition-and-attitude-determination-based-on-tactile-pressure-analysis-cnn-based-posture-determination/253015.png","ImageObject",442,249,{"name":42,"@type":43},"Berry Peter","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-30","2026-09-13",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",7,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"Why does the method avoid visual guidance during robotic grasping?","Question",{"text":63,"@type":64},"Existing robotic grasping relies on visual guidance, which is limited by environmental lighting and object occlusion. For fine operations, macroscopic vision is not suitable.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"How is the tactile contact data converted for learning?",{"text":68,"@type":64},"The approach constructs a contact surface deformation model and performs dimensionality reduction on vector force arrays by decomposing the 3D contact surface into separate force planes. It outputs a 2D scalar array containing force magnitude for CNN input.",{"name":70,"@type":61,"acceptedAnswer":71},"What performance does the proposed CNN-based method achieve?",{"text":72,"@type":64},"Experiments show millimeter-level precision for grasp position determination and a two-dimensional posture angle error of less than 1 degree. This generally meets requirements for power system business scenarios.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},253015,1790363548,{"code":4,"msg":82,"data":83},"success",[84,88,93,98,103,108,113,118,123],{"id":85,"doc_module":22,"doc_module_name":25,"category_name":29,"show_sort_weight":86,"slug":87},11,90,"presentations",{"id":89,"doc_module":22,"doc_module_name":25,"category_name":90,"show_sort_weight":91,"slug":92},12,"Resumes",80,"resumes",{"id":94,"doc_module":22,"doc_module_name":25,"category_name":95,"show_sort_weight":96,"slug":97},14,"Invoices",70,"invoices",{"id":99,"doc_module":22,"doc_module_name":25,"category_name":100,"show_sort_weight":101,"slug":102},15,"Posters",60,"posters",{"id":104,"doc_module":22,"doc_module_name":25,"category_name":105,"show_sort_weight":106,"slug":107},16,"Social Media",50,"social-media",{"id":109,"doc_module":22,"doc_module_name":25,"category_name":110,"show_sort_weight":111,"slug":112},17,"Forms",40,"forms",{"id":114,"doc_module":22,"doc_module_name":25,"category_name":115,"show_sort_weight":116,"slug":117},18,"Letters",30,"letters",{"id":119,"doc_module":22,"doc_module_name":25,"category_name":120,"show_sort_weight":121,"slug":122},21,"Paper Templates",5,"papers-templates",{"id":124,"doc_module":22,"doc_module_name":25,"category_name":125,"show_sort_weight":4,"slug":126},158,"General","general-158",{"code":4,"msg":82,"data":128},{"doc_id":79,"user_id":129,"nickname":42,"user_avatar":130,"doc_module":22,"category_id":85,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":131,"file_id":132,"file_url":133,"file_type":134,"file_size":135,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":104,"language":136,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":137,"faqs":138,"seo_title":139,"seo_description":12,"update_tm":140,"read_time":141},1374402524268,"https://ap-avatar.wpscdn.com/davatar_3d24733baf745e90a7e4bdd5f77d97b2","[www. nature.com/scientificreports](www. nature.com/scientificreports)  \nOPEN  \nResearch on blindsight technology for object recognition and attitude determination based on tactile pressure analysis  \nChen Li, Li HuiJun􀀍 & Song Aiguo  \nAiming at the existing robotic hand’s reliance on visual guidance for object recognition during the grasping process, which is limited by environmental lighting and object occlusion, this paper is committed to proposing a method for object recognition and posture determination based on contact force feedback. The paper first constructs a contact surface deformation model for classic button operations and grasping operations in power systems, based on Hertz contact and elastoplastic unloading theories, laying the theoretical foundation for the vector mechanics decomposition of the contact surface. It then proposes a dimensionality reduction analysis method for vector force arrays, identifying the boundary force characteristics of the contact surface to decompose the 3D contact surface into separate force planes and reducing the angle dimension in the vector array to a twodimensional scalar array that only includes the magnitude of the force, thus laying the data foundation for introducing convolutional neural networks (CNNs). Through training and learning the mechanical features of the contact surface, the paper achieves perception and posture determination of objects during pressing and grasping processes. Practical experiments demonstrate that the proposed method for rapid analysis and extended posture determination of the grasping surface based on CNNs can achieve millimeter-level precision in grasping position determination and a two-dimensional posture angle of less than 1 degree for the grasped object, generally meeting the requirements of power system business scenarios. The blind-sense technology explored in this paper can be widely applied to non-visual perception at the end of humanoid robots and has significant engineering guidance value.  \nKeywords Visual guidance, Vector force decomposition, Blind sensing technology, Attitude determination  \nSoft grasping is the foundation of a robotic hand’s precision grasping. Unlike rigid contact grasping, soft grasping produces varying degrees of deformation on the contact surface. Existing mechanical detection methods are unable to accurately identify the posture of the grasped object and the stability of the contact surface. Macroscopic visual acquisition outside the robotic hand’s sensors can, to some extent, obtain the posture of the grasped object, but it is limited by on-site lighting conditions and occlusions. During fine operations, macroscopic visual detection is not suitable. Therefore, scholars have conducted extensive research on the stability of grasping and posture determination, including the generation of datasets, grasping algorithms, multimodal perception, functional and adaptive grasping, and real-time interaction.  \nReference1 provides an efficient framework for synthesizing diverse and stable grasping posture data, generating the large-scale simulation dataset DexGraspNet, which significantly enhances the performance of dexterous robotic hand grasping synthesis algorithms. Reference2 divides the grasping process into static grasping gesture generation and reinforcement learning-based execution of grasping based on target gestures, allowing the algorithm to generalize across various objects and adapt to downstream tasks. Reference3 proposes strategies for handling objects of various shapes and features, adapting to changes in geometric shapes, textures, or lighting conditions, and demonstrating extensive adaptability to a wide range of objects. Additionally, Reference4 achieves functional dexterous grasping for robotic arms through a human-to-robot grasping redirection module, transferring human grasping postures to different robotic hands. Reference5 combines vision-language-action models with diffusion models to enhance the ","cbCaiveCKecvVcWi","https://ap.wps.com/l/cbCaiveCKecvVcWi","pdf",3216954,"English","# Research on Blindsight Technology for Object Recognition and Attitude Determination Based on Tactile Pressure Analysis\n## Tactile-force-based modeling and vector decomposition\n## CNN-based recognition and posture determination\n## Experimental results and application value","[{\"question\":\"Why does the method avoid visual guidance during robotic grasping?\",\"answer\":\"Existing robotic grasping relies on visual guidance, which is limited by environmental lighting and object occlusion. For fine operations, macroscopic vision is not suitable.\"},{\"question\":\"How is the tactile contact data converted for learning?\",\"answer\":\"The approach constructs a contact surface deformation model and performs dimensionality reduction on vector force arrays by decomposing the 3D contact surface into separate force planes. It outputs a 2D scalar array containing force magnitude for CNN input.\"},{\"question\":\"What performance does the proposed CNN-based method achieve?\",\"answer\":\"Experiments show millimeter-level precision for grasp position determination and a two-dimensional posture angle error of less than 1 degree. This generally meets requirements for power system business scenarios.\"}]","Research on Blindsight Technology for Object Recognition and Attitude Determination Based on Tactile Pressure Analysis - CNN-based posture determination | PDF",1789263816,6]