[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-81720-en":3,"doc-seo-81720-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},81720,4810365810221,"Aurora","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","FLYNN: Robust Neural Network for Robot Navigation using Fly Brain Topology","Deep learning models deliver strong performance yet remain brittle under new environments or missing sensory inputs. Biological systems, by contrast, tolerate such challenges. This work develops a recurrent neural network whose architecture is derived from the synaptic-resolution brain connectome of Drosophila melanogaster. The fly connectome neural network (FLYNN) is trained for vision-based navigation in MuJoCo, matching modern hand-crafted networks in parameter count, while showing enhanced robustness to out-of-distribution data and sensory loss, with functionality persisting under total vision loss.","FLYNN: Robust Neural Network for Robot Navigation using Fly Brain  \nTopology  \nBenquan Wang 1 and Jingdao Chen2  \narXiv :2607 .00025v1 [ cs .RO] 21 Jun 2026  \nAbstract—While deep learning models achieve state-of-theart performance in complex tasks, they remain brittle when faced with new environments or sensory deprivation. In contrast, biological systems exhibit remarkable tolerance to these challenges. We address this vulnerability by developing arecurrent neural network (RNN) whose architecture is directly derived from the synaptic-resolution brain connectome of the fruit fly Drosophila melanogaster. We demonstrate the feasibility of training the fly connectome neural network (FLYNN) to perform vision-based navigation in MuJoCo, achieving performance comparable to modern hand-crafted networks of similar parameter counts. Crucially, FLYNN exhibits superior resistance to out-of-distribution (OOD) data and tolerance to sensory loss without further training. It remained functional even under total vision loss while hand-crafted networks largely failed, even when specifically trained with camera dropout. Principal Component Analysis (PCA) of the internal state of FLYNN suggests that it exhibits a particularly high degree of representational modularity, which might be related to its robustness. Our work provides a new direction for designing resilient artificial agents following the topology of biological brains.  \nI. INTRODUCTION  \nArtificial Neural Networks (ANNs) originated from biologically inspired artificial neurons [1] . Since their inception, researchers have engineered increasingly sophisticated architectures to address complex challenges such as voice recognition, image classification, and navigation. Although the detailed structures of modern ANNs are primarily handcrafted, the biological brain remains a foundational source of inspiration. The Convolutional Neural Network (CNN)  \n[2], [3] is a prominent example of this influence, followed by other notable architectures including Recurrent Neural Networks (RNNs) [4], Long Short-Term Memory (LSTM) units [5], and Transformers [6] . Following decades of development, state-of-the-art ANN architectures now frequently outperform humans on a wide range of complex tasks.  \nNavigation is one of the areas in which robustness is critical. While current state-of-the-art ANN models perform well under ideal conditions, they are known to be vulnerable to out-of-distribution (OOD) data [7] and sensor information degradation or deprivation. In modern autonomous driving systems, for example, encountering a never-seen-before situation or suffering the loss of a camera stream typically triggers an immediate emergency handover, as the underlying  \n1Benquan Wang is with the Department of Computer Science Engineering, Mississippi State University, Mississippi State, MS 39762, USA [bw1918@msstate.edu](bw1918@msstate.edu)  \n2Jingdao Chen is with the Department of Computer Science Engineering, Mississippi State University, Mississippi State, MS 39762, USA [chenjingdao@cse.msstate.edu](chenjingdao@cse.msstate.edu)  \nmodels are designed to rely on the continuous integration of all available inputs [8] according to their trained policies.  \nIn contrast, biological organisms exhibit extraordinary robustness to all of these failure modes. For example, animals are known to adapt to new environments without difficulty. An animal that loses an eye can typically adapt quickly and maintain its ability to navigate the environment. Such resilience is likely rooted in the distinctive architecture of the biological brain. In Drosophila, for instance, the visual system is composed of distinct, highly organized neuropils – including the lamina, medulla, lobula plate, and lobula – specialized for calculating optical flow from visual input [9],[10] . Furthermore, the central complex of the Drosophila brain contains a population of ring neurons that serve as a heading direction calculator for navigation [11], [12]","cbCaiinkTUUaAvSu","https://ap.wps.com/l/cbCaiinkTUUaAvSu","pdf",5540911,2,1,9,"English","en",105,"# Introduction\n## Motivation: ANN brittleness and navigation robustness\n## Biological prior: Drosophila visual system and heading direction coding\n## Related work and remaining challenges\n## Our approach and evaluation setup","[{\"question\":\"What problem does FLYNN address in robot navigation?\",\"answer\":\"FַLYNN targets brittleness of deep models when facing out-of-distribution environments or degraded and missing sensory inputs during navigation.\"},{\"question\":\"How is the FLYNN architecture constructed?\",\"answer\":\"FLYNN is implemented as a recurrent neural network whose connectivity is directly derived from the synaptic-resolution brain connectome of the fruit fly Drosophila melanogaster.\"},{\"question\":\"What robustness behaviors does FLYNN demonstrate compared with hand-crafted networks?\",\"answer\":\"FLYNN shows superior resistance to out-of-distribution data and tolerance to sensory loss without further training, remaining functional even under total vision loss where hand-crafted networks largely fail.\"}]",1784175627,23,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"flynn-robust-neural-network-for-robot-navigation-using-fly-brain-topology","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,47,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":20},"https://docshare.wps.com/document/","Document",{"item":48,"name":12,"@type":43,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/flynn-robust-neural-network-for-robot-navigation-using-fly-brain-topology/81720/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-23","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does FLYNN address in robot navigation?","Question",{"text":75,"@type":76},"FַLYNN targets brittleness of deep models when facing out-of-distribution environments or degraded and missing sensory inputs during navigation.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How is the FLYNN architecture constructed?",{"text":80,"@type":76},"FLYNN is implemented as a recurrent neural network whose connectivity is directly derived from the synaptic-resolution brain connectome of the fruit fly Drosophila melanogaster.",{"name":82,"@type":73,"acceptedAnswer":83},"What robustness behaviors does FLYNN demonstrate compared with hand-crafted networks?",{"text":84,"@type":76},"FLYNN shows superior resistance to out-of-distribution data and tolerance to sensory loss without further training, remaining functional even under total vision loss where hand-crafted networks largely fail.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,127,130,134],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]