[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85692-en":3,"doc-seo-85692-105":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},85692,137441390410,"Hazel","https://ap-avatar.wpscdn.com/avatar/2000252f4ab5702993?_k=1776741390130283984",8,"Research & Report","A Dynamic Scene Interaction Reasoning Framework for Scene-level Lane-Change Intention and Trajectory Prediction of Multiple Interacting Vehicles","Safe motion planning for ADAS and autonomous vehicles depends on accurately forecasting how surrounding traffic scenes evolve. Existing work often focuses on a single target vehicle for lane-change intention and trajectory prediction, while multi-agent forecasting may provide limited explicit maneuver information per agent. This study proposes a dynamic scene graph attention framework that predicts lane-change intention and future trajectories for all relevant vehicles. Vehicles form a time-varying interaction graph with explicit spatial and kinematic edge features, enabling temporal graph-attention message passing and an intention-guided decoder, plus scene-level consistency. Experiments on NGSIM I-80, NGSIM US-101, and highD show consistent gains, higher intention accuracy and lower RMSE, collision rates, and joint displacement errors.","A Dynamic Scene Interaction Reasoning Framework for Scene-level Lane-Change Intention and Trajectory Prediction of Multiple Interacting Vehicles⋆  \narXiv :2607 .09740v 1 [ cs .AI] 2 Jul 2026  \nJoshua Kofi Asamoaha,1 , Blessing Agyei Kyema,2 , Eugene Denteha,3 and Armstrong Aboaha,∗  \na North Dakota State University, 1410 14th Avenue Offerdahl North Building, CIE 201, Fargo, 58102, North Dakota, United States  \nARTICLE INFO  \nKeywords:  \nlane-change intention prediction trajectory forecasting dynamic scene graph  \ngraph attention  \nautonomous driving  \nAB STRACT  \nSafe motion planning in advanced driver-assistance systems and autonomous vehicles requires an accurate understanding of how the surrounding traffic scene is likely to evolve. However, many existing lane-change prediction methods remain centered on a single target vehicle, while multiagent forecasting approaches often describe scene evolution only through future positions and provide limited explicit information about the maneuver associated with each vehicle. This study proposesa dynamic scene graph attention framework that predicts the lane-change intention and future trajectory of every relevant vehicle within a local traffic scene. The scene is represented as a timevarying interaction graph in which vehicles are modeled as nodes and their spatial and kinematic relationships are encoded through explicit edge features. Temporal graph-attention message passing captures evolving inter-vehicle dependencies and pre-maneuver cues, while an intention-guided decoder links each predicted maneuver to its corresponding future motion. A scene-level consistency objective further encourages compatible multi-vehicle futures. Experiments on the NGSIM I-80, NGSIM US-101, and highD datasets demonstrate consistent improvements over competing baselines. DSiGAT achieves intention prediction accuracies of 90.12% and 90.97% on NGSIM I-80 and US-101, respectively, and reduces trajectory RMSE by up to 52.94% relative to the strongest baseline. It also produces lower inter-agent collision rates and joint displacement errors, indicating more coherent scene-level predictions. Ablation, sensitivity, robustness, and qualitative analyses further validate the contribution of the proposed components and the effectiveness of the scene-focused formulation.  \n1. Introduction  \nAdvanced driver-assistance systems (ADAS) and autonomous vehicles (AVs) are expected to anticipate how the surrounding traffic environment is likely to evolve in order to make safe and timely driving decisions Sharma and Awasthi (2022) . Although perception systems describe the current traffic scene Asamoah et al. (2026), planning requires an understanding of how that scene may changeover the next few seconds Hagedorn et al. (2024); Kim et al.(2021) . This is challenging in mixed traffic, where several nearby vehicles may act simultaneously and influence the actions available to the ego vehicle. A prediction focused on only one selected vehicle can provide only a partial view of the future traffic situation. Accordingly, reliable planning requires the behavior of the reference vehicle and all relevant surrounding vehicles to be considered within the same local  \n⋆  \n∗Corresponding Author  \n [joshua.asamoah@ndsu.edu](joshua.asamoah@ndsu.edu) (J.K. Asamoah); [blessing.agyeikyem@ndsu.edu](blessing.agyeikyem@ndsu.edu) (B.A. Kyem); [eugene.denteh@ndsu.edu](eugene.denteh@ndsu.edu) (E. Denteh); [armstrong.aboah@ndsu.edu](armstrong.aboah@ndsu.edu) (A. Aboah)  \n [https://joshuakasamoah.github.io/](https://joshuakasamoah.github.io/) (J.K. Asamoah); [https://blessing-agyei-kyem.github.io/](https://blessing-agyei-kyem.github.io/) (B.A. Kyem); [https://aboaharmstrong.vercel.app/](https://aboaharmstrong.vercel.app/) (A. Aboah)  \nORCID(s): 0009-0002-3258-0479 (J.K. Asamoah); 0009-0006-6360-6386 (B.A. Kyem); 0009-0001-7597-6732 (E. Denteh); 0000-0002-1605-1545 (A. Aboah)  \n1  \ntraffic scene Joseph and Mondal (2021); Schulz et al. (2018); Bente","cbCaigaN4O1q3SeB","https://ap.wps.com/l/cbCaigaN4O1q3SeB","pdf",1583238,1,28,"English","en",105,"# Introduction\n## Problem motivation and limitations of target-centered prediction\n## Lane change as a distributed multi-vehicle response\n## Related approaches and need for scene-level multi-agent modeling","[{\"question\":\"What limitation affects many existing lane-change prediction methods?\",\"answer\":\"Many methods are formulated around a predefined target vehicle, using surrounding vehicles only as context without predicting their future behaviors in the same scene.\"},{\"question\":\"How does the proposed framework represent the traffic scene?\",\"answer\":\"It represents the local scene as a time-varying interaction graph where vehicles are nodes, and spatial/kinematic relationships are encoded via explicit edge features.\"},{\"question\":\"What results demonstrate the effectiveness of the method?\",\"answer\":\"Experiments on NGSIM I-80, NGSIM US-101, and highD show improved intention prediction accuracy (90.12% and 90.97%) and up to 52.94% lower trajectory RMSE versus the strongest baseline, along with reduced inter-agent collision rates and joint displacement errors.\"}]",1784205633,71,{"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},"a-dynamic-scene-interaction-reasoning-framework-for-scene-level-lane-change-intention-and-trajectory-prediction-of-multiple-interacting-vehicles","",{"@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/a-dynamic-scene-interaction-reasoning-framework-for-scene-level-lane-change-intention-and-trajectory-prediction-of-multiple-interacting-vehicles/85692/",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","2026-07-17","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 limitation affects many existing lane-change prediction methods?","Question",{"text":75,"@type":76},"Many methods are formulated around a predefined target vehicle, using surrounding vehicles only as context without predicting their future behaviors in the same scene.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the proposed framework represent the traffic scene?",{"text":80,"@type":76},"It represents the local scene as a time-varying interaction graph where vehicles are nodes, and spatial/kinematic relationships are encoded via explicit edge features.",{"name":82,"@type":73,"acceptedAnswer":83},"What results demonstrate the effectiveness of the method?",{"text":84,"@type":76},"Experiments on NGSIM I-80, NGSIM US-101, and highD show improved intention prediction accuracy (90.12% and 90.97%) and up to 52.94% lower trajectory RMSE versus the strongest baseline, along with reduced inter-agent collision rates and joint displacement errors.","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"]