[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-81932-en":3,"doc-seo-81932-105":29,"detail-sidebar-cat-0-en-105":90},{"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":13,"seo_description":14,"update_tm":27,"read_time":28},81932,8796095461564,"Liam","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","TypeGo: An OS Runtime for Embodied Agents","Large language models can plan behavior for embodied agents from natural language, yet using the LLM as an on-path request/response oracle conflicts with real-time control and concurrent goals. The work proposes an operating-system-style runtime and instantiates it in the TypeGo prototype. TypeGo organizes LLM planning as overlapping asynchronous loops at multiple timescales, arbitrates physical subsystems via a Skill Kernel, and hides inference latency through speculative skill streaming plus a fast first-action path. A Kalos quadruped prototype shows reduced step delay and faster time-to-first-action while supporting concurrent tasks with low scheduling overhead.","TypeGo: An OS Runtime for Embodied Agents  \nGuojun Chen  \n[guojun.chen@yale.edu](guojun.chen@yale.edu)[ ](guojun.chen@yale.edu)Yale University  \nAlex Schott  \n[a.schott@yale.edu](a.schott@yale.edu)[ ](a.schott@yale.edu)Yale University  \nLin Zhong  \n[lin.zhong@yale.edu](lin.zhong@yale.edu)[ ](lin.zhong@yale.edu)Yale University  \narXiv :2607 .05482v 1 [ cs . SE] 6 Jul 2026  \nAbstract  \nLarge language models (LLMs) can plan behavior for embodied agents from natural language, but treating the LLM as a request/response oracle on the critical path is fundamentally at odds with real-time control and concurrent goals. We argue for an operatingsystem-style runtime for embodied agents, and instantiate this idea in an early prototype, TypeGo. TypeGo structures LLM-based planning as asynchronous loops at multiple timescales that overlap with execution, and manages the agent’s physical body like an OS manages hardware: the Skill Kernel arbitrates typed physical subsystems among concurrent per-task processes, a scheduler preempts them and resumes or replaces each by source, and speculative skill streaming hides LLM latency behind ongoing motion, while a fast first-action path yields visible feedback within a second. Users program behavior through natural language prescriptions that TypeGo dispatches to the LLM-based planners or compiles into lowlatency interrupt handlers. Our prototype ofKalos, a Unitree Go2 quadruped, provides preliminary evidence for the design: in our current task suite, it cuts per-step delay by 50% over step-by-step planning and time-to-first-action by 73% over monolithic planning, while admitting concurrent tasks at low scheduling overhead.  \n1 Introduction  \nRobots today execute a rich catalog of low-level skills: quadrupeds walk, trot, climb stairs, and balance, while humanoids add grasping and manipulation on top; many platforms ship turnkey APIs for locomotion, speech, and perception. These skills solve basic motion and action well, but they do not tell a robot what to do when a human asks for something new, the scene changes unexpectedly, or several goals compete for attention. The hard problem has shifted from how a robot body moves to how an embodied system decides, and hard-coded task logic or per-task trained policies do not scale to the open world.  \nLLMs offer a path to close the decision-making gap: pretrained on large-scale data, they supply the common sense to interpret a situation and compose the robot’s existing skills into a plan [1, 2] . But using an LLM to drive an embodied agent raises three system challenges that persist even as models become faster and more capable.  \n(i) First, LLM inference is too slow for real-time reaction. There are systems that place a language or vision–language model in the perception–action loop for high-level reasoning to achieve impressive open-world generality [3–7] . But they operate far below real-time, with long pauses between consecutive actions, so demonstrations are commonly shown at increased playback speed. Small models respond fast but plan poorly, while state-of-the-art models are capable but substantial, adding seconds of latency. This latency– quality tradeoff is fundamental, and neither extreme alone suffices for real-time control. (ii) Second, real-world embodied agents rarely  \nrun one task in isolation. It may be patrolling an area while communicating with a nearby human, and still needs to watch for hazards. Arbitrating the robot’s shared actuators among concurrent tasks while honoring their priorities and preempting cleanly are classic OS problems that the LLM literature has largely sidestepped. (iii) Third, one-shot natural-language instructions cover only goals. Embodied agents also need reactive rules (e.g.,“back up if anything show abruptly within a near space (0 . 2m)”) that must fire far faster than an LLM call. Making natural language a first-class medium for both goals and fast reaction rules, authored by non-programmers without a code editor","cbCairaVayWaLF3s","https://ap.wps.com/l/cbCairaVayWaLF3s","pdf",535841,1,7,"English","en",105,"# Introduction\n## Core challenge: real-time and concurrent embodied control\n## Runtime hypothesis: continuous asynchronous execution\n## Three mechanisms\n# Multi-cadence asynchronous planning\n# OS-style runtime with semantic scheduling\n# Natural language as a first-class programming medium","[{\"question\":\"Why is an LLM request/response loop insufficient for embodied agents?\",\"answer\":\"LLM inference introduces latency that conflicts with real-time reaction, and embodied agents must handle multiple concurrent tasks while arbitrating shared actuators. The paper argues that a continuous asynchronous runtime avoids placing the LLM on the critical path.\"},{\"question\":\"How does TypeGo structure planning to handle latency and adaptivity?\",\"answer\":\"TypeGo runs multiple concurrent planning loops at different timescales, including a fast reflex layer and an action streamer. Speculative skill streaming overlaps planning with skill execution and buffers upcoming skills in a bounded queue to hide LLM latency behind motion.\"},{\"question\":\"How does TypeGo manage concurrent tasks and shared robot resources?\",\"answer\":\"TypeGo assigns tasks to processes and uses a Skill Kernel to arbitrate categorized physical subsystems. The scheduler uses semantic scheduling to decide which process wins when resources conflict, and defines whether preempted tasks resume based on their origin.\"}]",1784177125,18,{"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":85,"head_meta":87,"extra_data":89,"updated_unix":27},"typego-an-os-runtime-for-embodied-agents","",{"@graph":35,"@context":84},[36,53,67],{"@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/typego-an-os-runtime-for-embodied-agents/81932/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"Why is an LLM request/response loop insufficient for embodied agents?","Question",{"text":74,"@type":75},"LLM inference introduces latency that conflicts with real-time reaction, and embodied agents must handle multiple concurrent tasks while arbitrating shared actuators. The paper argues that a continuous asynchronous runtime avoids placing the LLM on the critical path.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How does TypeGo structure planning to handle latency and adaptivity?",{"text":79,"@type":75},"TypeGo runs multiple concurrent planning loops at different timescales, including a fast reflex layer and an action streamer. Speculative skill streaming overlaps planning with skill execution and buffers upcoming skills in a bounded queue to hide LLM latency behind motion.",{"name":81,"@type":72,"acceptedAnswer":82},"How does TypeGo manage concurrent tasks and shared robot resources?",{"text":83,"@type":75},"TypeGo assigns tasks to processes and uses a Skill Kernel to arbitrate categorized physical subsystems. The scheduler uses semantic scheduling to decide which process wins when resources conflict, and defines whether preempted tasks resume based on their origin.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,118,121,126,129,133],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":21,"doc_module":4,"doc_module_name":45,"category_name":115,"show_sort_weight":116,"slug":117},"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":119,"slug":120},30,"research-report",{"id":122,"doc_module":4,"doc_module_name":45,"category_name":123,"show_sort_weight":124,"slug":125},9,"Religion & Spirituality",20,"religion-spirituality",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":124,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":45,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":45,"category_name":135,"show_sort_weight":105,"slug":136},19,"General","general"]