[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86169-en":3,"doc-seo-86169-105":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},86169,962075114101,"Seraphina","https://ap-avatar.wpscdn.com/avatar/e000253a75eb197efd?x-image-process=image/resize,m_fixed,w_180,h_180&k=1780044092746381165",8,"Research & Report","Stop to Decide: Latency-Aware Proprioceptive Navigation Primitives for Mapping-Free Quadruped Inspection","Compute-constrained quadrupeds often run their navigation loop well below the controller’s design rate, and shared onboard vision compute can reduce a stair loop to about 15 Hz. This latency disrupts a standard proprioceptive stair rule, causing overshoot of the step edge during climbing. On a stepped platform, a climb–settle cadence keeps overshoot near zero at each loop rate, while a logistic dose–response model explains failures versus per-period advance. A fully onboard mapping-free, learning-free stack then performs line following, 90° corner maneuvers in a 55 cm corridor, and stair traversal, completing the inspection course in 18/20 trials (90%).","arXiv :2607 . 11204v1 [ cs .RO] 13 Jul 2026  \nStop to Decide: Latency-Aware Proprioceptive Navigation Primitives for Mapping-Free Quadruped Inspection  \nHanting Suo 1†, Haonan Yan2†, Liang Wang3 , Aiguo Song 1*  \n1* School of Instrument Science and Engineering, Southeast University, Nanjing, China.  \n2 School of Electrical Engineering, Southeast University, Nanjing, China.  \n3 School of Mechanical Engineering, Southeast University, Nanjing, China.  \n*Corresponding author(s). E-mail(s): [a.g.song@seu.edu.cn](a.g.song@seu.edu.cn) ; Contributing authors: [213233673@seu.edu.cn](213233673@seu.edu.cn) ; [213230424@seu.edu.cn](213230424@seu.edu.cn) ; wl [seu@seu.edu.cn](seu@seu.edu.cn) ;  \n†These authors contributed equally to this work.  \nAbstract  \nCompute-constrained quadrupeds often run their navigation loop far below the controller’s design rate: sharing the onboard Jetson Orin with the vision pipeline slows our stair loop to ≈15 Hz. This latency breaks a standard proprioceptive pattern — declaring stair-summit arrival from the bodypitch signal while still climbing. On a stepped platform whose 50 cm top is shorter than the robot (Unitree Go2, ≈75 cm), in-motion detection overshoots the top edge with probability rising with the per-period advance v/f (the slowest ≈15 Hz cell partly diluted by a separate non-arrival mode), whereas a climb–settle cadence holds overshoot near zero at every loop rate (pooled 22/45 vs 1/45 over ≈30/20/15 Hz; Fisher p ≈ 2.4 × 10 −7; 7/15 vs 0/15 at the deployed ≈15 Hz) . A logistic dose–response model in v/f captures the failure; a pre-specified 40 Hz out-of-sample test favours the protocol-clean fit (33% observed vs 43%/22% predicted), giving a deployment rule (critical loop rate ≈19 Hz at 0.30 m/s) . The detector sits in a fully onboard, mapping-free and learning-free stack  \n—built-in inertial measurement unit, four foot-force channels, three 1-D ranges, one line camera—chaining line-following, a three-segment maneuver for 90◦ corners in a 55 cm corridor (20/20 contactfree vs 14/20 with 12 wall contacts for in-place yaw; exit-heading error 1.56 ◦ vs 5.64◦ ), and stair traversal, completing the inspection course in 18/20 trials (90%) . Results are from a single course geometry, platform, and operator.  \nKeywords: quadruped robot, stair traversal, control-loop latency, proprioceptive sensing, narrow-corridor maneuver, inspection robotics  \n1 Introduction  \nIndustrial inspection has become one of the most-cited deployment contexts for legged robots. Quadrupeds such as the ANYmal X and the Boston Dynamics Spot are now in routine commercial use at offshore platforms, oil-and-gas terminals, geothermal plants, electrical converter  \nstations, and fusion-research facilities, where the recurring operational profile is repetitive traversal of stairs, grating walkways, narrow service corridors, and painted or taped guide lines between equipment cabinets [1–7] . Surveys of robotic inspection in buildings, construction, and infrastructure consistently identify sensor cost, payload, and perception-stack computational complexity as  \nthe dominant adoption barriers [8, 9] . These barriers are most acute when the deployment target isnot a frontier research platform but a commercial quadruped on which the engineering team must build a reproducible, onboard, low-maintenance autonomy stack.  \nA natural question is whether the perceptive sensing layer typically assumed by these autonomy stacks is strictly necessary for the structuredinspection segment of the deployment envelope. Recent quadruped-navigation systems aimed atthe same hardware regime—for example the ROS 2 + SLAM-Toolbox + Nav2 pipelines reported on Spot and on the Unitree Go2 EDU platform [10]  \n— rely on full 3D LiDAR point clouds, RGB-D depth, or fused multi-modal perception to localize and to detect terrain transitions. In parallel, a substantial body of locomotion-level work has shown that proprioceptive signals alone — joint state, body-mounted IMU, an","cbCaip5LKw62Y7Ab","https://ap.wps.com/l/cbCaip5LKw62Y7Ab","pdf",3359483,3,1,30,"English","en",105,"# Introduction\n## Related Work and Problem Motivation\n## Research Question and Contribution Overview\n## System Setup and Sensing Budget","[{\"question\":\"Why does control-loop latency cause failures in proprioceptive stair navigation?\",\"answer\":\"When the navigation loop runs far below the intended controller rate, the robot’s proprioceptive stair-arrival declaration can occur while it is still climbing, leading to edge overshoot that increases with per-period advance v/f.\"},{\"question\":\"What strategy reduces overshoot when operating under low loop rates?\",\"answer\":\"A climb–settle cadence maintains overshoot near zero across loop rates, outperforming the standard proprioceptive pattern that declares arrival from body-pitch during ascent.\"},{\"question\":\"How does the proposed robot complete the inspection course without mapping or learning?\",\"answer\":\"The stack is fully onboard and uses only IMU signals, foot-force contact sensors, 1-D distance ranges, and monocular line-following for the painted centerline, chaining line following, corner maneuvers, and stair traversal to finish in 18/20 trials.\"}]",1784209072,76,{"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},"stop-to-decide-latency-aware-proprioceptive-navigation-primitives-for-mapping-free-quadruped-inspection","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,50],{"item":41,"name":42,"@type":43,"position":21},"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":20},"https://docshare.wps.com/document/research-report/",{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/stop-to-decide-latency-aware-proprioceptive-navigation-primitives-for-mapping-free-quadruped-inspection/86169/",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","2026-07-26","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},"Why does control-loop latency cause failures in proprioceptive stair navigation?","Question",{"text":75,"@type":76},"When the navigation loop runs far below the intended controller rate, the robot’s proprioceptive stair-arrival declaration can occur while it is still climbing, leading to edge overshoot that increases with per-period advance v/f.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What strategy reduces overshoot when operating under low loop rates?",{"text":80,"@type":76},"A climb–settle cadence maintains overshoot near zero across loop rates, outperforming the standard proprioceptive pattern that declares arrival from body-pitch during ascent.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the proposed robot complete the inspection course without mapping or learning?",{"text":84,"@type":76},"The stack is fully onboard and uses only IMU signals, foot-force contact sensors, 1-D distance ranges, and monocular line-following for the painted centerline, chaining line following, corner maneuvers, and stair traversal to finish in 18/20 trials.","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},[93,97,101,105,110,115,120,122,127,130,134],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"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":22,"slug":121},"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"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"]