[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-125311-en":3,"doc-seo-125311-105":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":4,"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":27,"seo_description":14,"update_tm":28,"read_time":29},125311,1374391974468,"Eden","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","Enhancing traffic dynamics-induced machine learning through heterogeneous driving policies - Research summary","Transportation systems increasingly require distributed computing to support real-time intelligent applications driven by large numbers of mobile agents. Conventional centralized cloud approaches create latency and bandwidth bottlenecks, while mobile edge/vehicular fog computing still depends on in-vehicle computational units for complex ML. This work proposes physical reservoir computing that exploits vehicle-fleet dynamics on roadways, projecting inputs to lead-vehicle speed, extracting nonlinear speed variation from following vehicles, and using a single-layer neural network for linear regression. Results show congestion level correlates with accuracy and heterogeneous driving policies improve performance, reaching a maximum accuracy of 0.787 with 30 vehicles.","npj | unconventional computing Article  \n\n| \u003Cbr>[https://doi.org/10.1038/s44335-025-00033-5](https://doi.org/10.1038/s44335-025-00033-5) |  |\n| --- | --- |\n| Enhancing traffic dynamics-induced machine learning through heterogeneous driving policies\u003Cbr> Check for updates |  |\n| Kai-Fung Chu1,2, Fan Ye1,2 , Arsen Abdulali1 & Fumiya Iida1 |  |\n| Transportation systems demand substantial computational resources to support diverse intelligent applications involving vast numbers of mobile agents atthe network edge. Existing approaches, such as mobile edge computing, merely redistribute computational tasks to edge devices, relying on invehicle computers as computational units. Here, we investigate an alternative computing approach: harnessing the inherent dynamics of physical vehicles without the need of in-vehicle computer as a complimentary and energy efficient computational resource. We propose a physical reservoir computing framework that can leverage dynamics produced by vehicle fleets on roadways and transform them into vast computational resources for various machine-learning (ML) tasks atthe edge of network. The proposed framework projects signal inputs to the lead vehicle speed to obtain anonlinear speed variation of the following vehicles, which is then used asthe system readout feeding toa single-layer neural network for linear regression based ona small amount of training data. Our results show a positive correlation between congestion level and accuracy, with heterogeneous driving policies significantly enhancing both. A heterogeneous fleet of 30 vehicles achieved a maximum accuracy of 0.787, comparable to an echo state network with 300 nodes, demonstrating excellent performance across various ML tasks. The results pave the way for utilizing mobile edge agents as novel physical computational resources for various ML tasks, which is crucial for enabling real-time computation close to the data source, significantly improving computing efficiency and latency for transport applications such as autonomous driving and traffic management. |  |\n| The rapid evolution of transportation systems calls for advanced computational resources and algorithms to meet the demands of intelligent, realtime applications involving countless mobile agents. From autonomous vehicle operations1 to real-time traffic management2 and intelligent traffic signal control3, modern transportation applications require swift, distributed computation4 that is often constrained by the limitations of centralized cloud infrastructures. Reliance on cloud servers creates bottlenecksin both communication bandwidth and latency, limiting response times in highly dynamic traffic environmentsand compromising the effectiveness of real-time systems. Mobile edge computing5 and vehicular fog computing6 has emerged as a partial solution, aiming to shift computation closer to the data source by redistributing tasks from the cloud to local vehicles. While these architectures reduce latency by enabling local data processing, it still heavily rely on computational resource at the edge such as an in-vehicle computer to process complex machine learning (ML) and decision-making | tasks. Consequently, there is a need to explore alternative or supplementary computational strategies that leverage unique characteristics of transportation systems to alleviate this constraint.\u003Cbr>Modern computational problems are frequently solved by siliconbased computers, whichhave achieved considerable success. However, these systems face significant drawbacks, including energy inefficiency and limitations in solving complex, scalable algorithms. For example, modeling physical dynamics in nature often requires solving partial differential questions, a task for which silicon-based computers are not specifically designed, leading to computationally intensive processes. In contrast, state transitionsin nature occur autonomously without intensive computation as required by silicon-based computers. By simply sensing ch","cbCainkCNUQAaBiz","https://ap.wps.com/l/cbCainkCNUQAaBiz","pdf",2147268,1,13,"English","en",105,"# Introduction\n## Unconventional and reservoir computing\n## Motivation from traffic prediction challenges\n## Proposed physical reservoir computing framework\n## Experimental results and implications","[{\"question\":\"What problem does the study address in transportation computing?\",\"answer\":\"It addresses the need for real-time distributed computation without relying on centralized cloud bottlenecks or expensive in-vehicle computers for complex ML tasks at the network edge.\"},{\"question\":\"How does the proposed physical reservoir computing framework work?\",\"answer\":\"It uses vehicle-fleet physical dynamics, projecting signal inputs to lead-vehicle speed to produce a nonlinear speed variation signal from following vehicles, then feeding this as a system readout to a single-layer neural network for linear regression.\"},{\"question\":\"What is the main finding about heterogeneous driving policies and accuracy?\",\"answer\":\"Accuracy increases with congestion level, and heterogeneous driving policies further enhance both the framework’s performance across tasks, achieving up to 0.787 with 30 vehicles.\"}]","Enhancing traffic dynamics-induced machine learning through heterogeneous driving policies - Research summary | PDF",1785898111,33,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"enhancing-traffic-dynamics-induced-machine-learning-through-heterogeneous-driving-policies-research-summary","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/enhancing-traffic-dynamics-induced-machine-learning-through-heterogeneous-driving-policies-research-summary/125311/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does the study address in transportation computing?","Question",{"text":75,"@type":76},"It addresses the need for real-time distributed computation without relying on centralized cloud bottlenecks or expensive in-vehicle computers for complex ML tasks at the network edge.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the proposed physical reservoir computing framework work?",{"text":80,"@type":76},"It uses vehicle-fleet physical dynamics, projecting signal inputs to lead-vehicle speed to produce a nonlinear speed variation signal from following vehicles, then feeding this as a system readout to a single-layer neural network for linear regression.",{"name":82,"@type":73,"acceptedAnswer":83},"What is the main finding about heterogeneous driving policies and accuracy?",{"text":84,"@type":76},"Accuracy increases with congestion level, and heterogeneous driving policies further enhance both the framework’s performance across tasks, achieving up to 0.787 with 30 vehicles.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"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":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"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":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]