[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83681-en":3,"doc-seo-83681-105":29,"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":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},83681,4810365810221,"Aurora","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Spiking Sequence Generator for Polar Trajectories on Neuromorphic Hardware","Neuromorphic controllers for SWaP-constrained systems demand neural architectures that are energy efficient and interpretable at the system-dynamics level. Prior work either uses end-to-end spiking networks with limited interpretability or converts classical controllers that underuse neuromorphic dynamics. A winner-take-all SNN with accessory populations is introduced to generate polar trajectories, including tuning rules for controlled neural transitions and shunting inhibition for independent direction, speed, and radius control. The network is implemented on SpiNNaker2, achieving two to three orders of magnitude lower step time and three to four orders lower energy use versus conventional computing.","A Spiking Sequence Generator for Polar Trajectories on  \nNeuromorphic Hardware  \nWilliam R.P. Nourse 1 ,2 ,∗ ORCID: 0000-0002-1437-026X  \nRoger D. Quinn 1 ORCID: 0000-0002-8504-7160  \narXiv :2607 .02753v 1 [ cs .NE] 2 Jul 2026  \n1 Mechanical & Aerospace Engineering, Case Western Reserve University, Cleveland, OH, USA  \n2 Mechanical & Aerospace Engineering, West Virginia University, Morgantown, WV, USA  \n∗ Corresponding author: [william.nourse@mail.wvu.edu](william.nourse@mail.wvu.edu) July 7, 2026  \nAbstract  \nNeuromorphic controllers for size, weight, and power-constrained systems require neural architectures that are both energy-efficient and interpretable at the level of system dynamics. However, existing approaches either rely on end-to-end trained spiking networks with limited interpretability, or on converted classical controllers that fail to fully exploit neuromorphic dynamics. We present a spiking neural network (SNN) architecture for generating polar trajectories, using a winner-take-all (WTA) architecture with accessory populations that induce controlled transitions in neural activity. We demonstrate tuning rules for these population dynamics, and utilize a form of shunting inhibition to enable independent control of direction, speed, and radius of the resulting polar trajectories. We implement the network on the SpiNNaker2 neuromorphic processor, and demonstrate a two to three orders of magnitude reduction in wall-clock step time and three to four orders of magnitude reduction in energy expenditure when compared to conventional computing platforms.  \nKeywords: winner-take-all; manifold; ring oscillator; neuromorphic engineering; event-based system; SpiNNaker2  \n1 Introduction  \nFor running neural controllers on size, weight, and power (SWaP) constrained robotic platforms, interest is growing in using brain-inspired neuromorphic computing hardware as a promising alternative to graphics processing units (GPUs), due to their low latency and high energy efficiency [1] . Work has been done to develop controllers for neuromorphic hardware using spiking neural networks (SNNs), with most work being done at one of two extremes: training deep artificial neural networks in an end-to-end manner [2, 3 , 4], or converting more classical control systems into SNNs which reproduce the desired dynamics [5, 6] . While these approaches are individually valid and sound, they come with tradeoffs for robotic applications. For end-to-end networks trained with gradient-based optimizers, questions of interpretability, safety, and stability arise as the networks usually behave as a black-box with little capability of human understanding. On the other extreme, converted classical controllers typically encounter approximation errors from discrete spike-timing, and provide minimal value when compared to modern low-power processors, which can evaluate classical control algorithms swiftly and efficiently. Ideally, networks deployed on neuromorphic hardware would offer a compromise somewhere in-between these two extremes. Neuromorphic  \nhardware is capable of generating rich temporal dynamics beyond traditional feedforward neural networks, and networks should be capable of utilizing these dynamics while allowing controllable interactions. These systems should also be able to use latent representations which are human interpretable, and compatible with feedback control. In total, the field needs a network solution which preserves low-dimensional interpretability, enables explicit control of trajectory dynamics, and maps efficiently to neuromorphic hardware.  \nWithin neuroscience, one approach which fulfills these criteria is that of neural manifolds [7] . The key insight of this area of work is that the primary variation in activity of large populations of neurons can often be reduced to a small number of degrees of freedom, using dimensionality reduction techniques such as principal component analysis. For neural circuits involved with behavior, t","cbCaishqNCqWqNFT","https://ap.wps.com/l/cbCaishqNCqWqNFT","pdf",1263903,1,21,"English","en",105,"# Introduction\n## Neural control on neuromorphic hardware\n## Limits of end-to-end SNN training and classical controller conversion\n## Neural manifolds and polar low-dimensional dynamics\n## Motivation for controllable low-dimensional SNN architectures","[{\"question\":\"What problem does the spiking sequence generator address for neuromorphic controllers?\",\"answer\":\"It targets the need for energy-efficient and interpretable neural architectures on SWaP-constrained robotic platforms, specifically for generating polar trajectory dynamics rather than relying on black-box end-to-end training or under-exploiting converted classical controllers.\"},{\"question\":\"How does the proposed network control polar trajectory parameters?\",\"answer\":\"It uses a winner-take-all (WTA) spiking architecture with accessory populations to induce controlled transitions, and a shunting inhibition mechanism that enables independent control of direction, speed, and radius.\"},{\"question\":\"What hardware implementation and performance gains are reported?\",\"answer\":\"The network is implemented on the SpiNNaker2 neuromorphic processor, showing a two to three orders of magnitude reduction in wall-clock step time and a three to four orders of magnitude reduction in energy expenditure compared with conventional computing platforms.\"}]",1784189707,53,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"spiking-sequence-generator-for-polar-trajectories-on-neuromorphic-hardware","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/spiking-sequence-generator-for-polar-trajectories-on-neuromorphic-hardware/83681/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-24","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 the spiking sequence generator address for neuromorphic controllers?","Question",{"text":75,"@type":76},"It targets the need for energy-efficient and interpretable neural architectures on SWaP-constrained robotic platforms, specifically for generating polar trajectory dynamics rather than relying on black-box end-to-end training or under-exploiting converted classical controllers.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the proposed network control polar trajectory parameters?",{"text":80,"@type":76},"It uses a winner-take-all (WTA) spiking architecture with accessory populations to induce controlled transitions, and a shunting inhibition mechanism that enables independent control of direction, speed, and radius.",{"name":82,"@type":73,"acceptedAnswer":83},"What hardware implementation and performance gains are reported?",{"text":84,"@type":76},"The network is implemented on the SpiNNaker2 neuromorphic processor, showing a two to three orders of magnitude reduction in wall-clock step time and a three to four orders of magnitude reduction in energy expenditure compared with conventional computing platforms.","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":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":45,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]