[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-81719-en":3,"doc-seo-81719-105":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},81719,4810365810221,"Aurora","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Urban Deceleration Behavior Modes Under Scene Context","Urban deceleration remains well-studied in car-following research yet lacks a clear taxonomy across different urban scene conditions. The study leverages Argoverse 2 multi-agent trajectory logs to extract 1,219 sustained deceleration events and represent each with a 19-dimensional kinematic feature vector. K-means discovers behavioral modes with bootstrap stability, while eleven scene-context variables quantify modulation. A HistGradientBoosting model predicts mode membership from the first 1.0 s, yielding stable modes and strong early-window predictive performance.","Urban Deceleration Behavior Modes Under Scene Context: An Early-Kinematic Classifier from Argoverse 2 Multi-Agent Trajectories  \nEni Solomon Laughter 1,*  \n1 School of Transportation Engineering, Chang’an University, Xi’an, Shaanxi, China; [2024134912@chd.edu.cn](2024134912@chd.edu.cn)  \n* Correspondence: [2024134912@chd.edu.cn](2024134912@chd.edu.cn);  \nAbstract: Urban deceleration is one of the most empirically studied yet least taxonomically organized behaviors in car-following research. With recent perception-equipped autonomous-vehicle datasets, enable trajectory-anchored mode discovery. We extract 1,219 sustained deceleration events from 234 urban driving logs of the Argoverse 2 Sensor dataset, encode each event in a 19-dimensional kinematic feature vector, discover behavioural modes via K-means clustering with bootstrap stability analysis, and quantify modulation by eleven scene-context variables. A HistGradientBoosting classifier predicts mode membership from the first 1.0 s of each event. Four stable modes emerge with bootstrap Adjusted Rand Index of 0.897 across 50 resamples —anticipatory soft (62 .8%), reactive closing (30.6%), brake-like jerk (4.8%), and an outlier category (1.8%). Only pair age shows a medium effect (ε² = 0.085); scene geometry and vulnerable-road-user proximity show negligible effects. The early-event classifier achieves macro-F1 = 0.758 at 1.0 s, with scene context contributing +0.059 F1 over kinematics alone. Modes are regime-invariant in medium-speed driving (ARI = 0.817) but regime-dependent at low speed (ARI = 0.166) . A small set of stable kinematic modes structures urban deceleration; early-window jerk dominates predictive signal; and pair age is the primary contextual modulator.  \nKeywords: deceleration behaviour; car-following; Argoverse 2; unsupervised clustering; early-event classification; scene context; advanced driver assistance systems  \n1. Introduction  \nWhat drivers do with their longitudinal control during when they drive, when they brake, how sharply, and in response to what visual and contextual cues, is one of the clearest behavioral signals captured by vehicle trajectory dataset. This control kinematic record is preserved at high temporal resolution, and the rapid maturation of perceptionequipped autonomous-vehicle datasets has shifted the empirical frontier from simulatorbased and instrumented-vehicle studies toward ground-truth-annotated multi-agent trajectory data. Where simulator studies offer controllability of stimuli and counterfactual replay, and where instrumented-vehicle naturalistic studies offer pedal-level fidelity, neither operates at the scale or scene diversity afforded by perception-AV datasets, in which dozens of co-present agents are tracked simultaneously with object class, geometry, and high-definition map context.  \nTrajectory data drawn from such datasets preserve the actual behavioral choices made by real drivers in real urban scenes, with measured rather than scripted contextual conditions, and without the behavioral changes that can occur when a driver knows they are operating an instrumented vehicle. Reaction-time microstructure, looming-driven brake initiation, and stop-phase transitions —all characterized under controlled or singlevehicle naturalistic conditions [1–3] manifest in trajectory data as observable kinematic  \nsignatures of behavioral modes. These properties motivate the present study’s use of the Argoverse 2 Sensor dataset [4] rather than simulator-generated or single-vehicle naturalistic data.  \nThe remainder of this introduction first lays out the conceptual background that underpins this study (Section 1.1) and then states the research objective and contributions to the growing body of trajectory-driving-behaviour work (Section 1.2) .  \n1.1. Background  \nCar-following modelling has progressed through fifty years of architectural refinement. From the constant-time-headway formulations of the 1950s, through the stimulus– response Gaz","cbCaiknjjzoTEYt1","https://ap.wps.com/l/cbCaiknjjzoTEYt1","pdf",1044931,3,1,23,"English","en",105,"# Abstract\n# Introduction\n## Background","[{\"question\":\"How are deceleration events represented and clustered in the study?\",\"answer\":\"The method encodes each sustained deceleration event from Argoverse 2 into a 19-dimensional kinematic feature vector, then uses K-means clustering with bootstrap stability analysis to discover behavioral modes.\"},{\"question\":\"How many deceleration modes are found and what characterizes them?\",\"answer\":\"Four stable modes emerge: anticipatory soft, reactive closing, brake-like jerk, and an outlier category. Their relative shares are quantified using bootstrap evaluation.\"},{\"question\":\"What factors most influence mode prediction from the early part of an event?\",\"answer\":\"A classifier predicts mode membership using only the first 1.0 s of each event. Early-event kinematic signals dominate the predictive strength, and scene context adds an incremental improvement; among contextual variables, pair age shows a medium effect while others are negligible.\"}]",1784175625,58,{"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},"urban-deceleration-behavior-modes-under-scene-context","",{"@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/urban-deceleration-behavior-modes-under-scene-context/81719/",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","2026-07-25","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},"How are deceleration events represented and clustered in the study?","Question",{"text":75,"@type":76},"The method encodes each sustained deceleration event from Argoverse 2 into a 19-dimensional kinematic feature vector, then uses K-means clustering with bootstrap stability analysis to discover behavioral modes.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How many deceleration modes are found and what characterizes them?",{"text":80,"@type":76},"Four stable modes emerge: anticipatory soft, reactive closing, brake-like jerk, and an outlier category. Their relative shares are quantified using bootstrap evaluation.",{"name":82,"@type":73,"acceptedAnswer":83},"What factors most influence mode prediction from the early part of an event?",{"text":84,"@type":76},"A classifier predicts mode membership using only the first 1.0 s of each event. Early-event kinematic signals dominate the predictive strength, and scene context adds an incremental improvement; among contextual variables, pair age shows a medium effect while others are negligible.","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},[93,97,101,105,110,115,120,123,128,131,135],{"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":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"]