[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84797-en":3,"doc-seo-84797-105":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},84797,5909877438554,"Maeve","https://ap-avatar.wpscdn.com/avatar/5600025385ad2bf12a7?_k=1778553567797529272",8,"Research & Report","VLM-CASE: Vision-Language Model Enabled Context-Adaptive Safety Envelopes for Anticipatory Safe Autonomous Driving","Adverse weather remains a core obstacle to autonomous driving because it simultaneously degrades both perception and vehicle maneuverability. The work introduces VLM-CASE, enabling anticipatory driving while keeping motion constrained by a formal safety model. A vision-language model, fine-tuned with LoRA, infers road surface and visibility from front-camera images to parameterize a context-adaptive safety envelope. The envelope couples braking and steering via a shared friction budget, and an MPC controller operates within it while VLM runs asynchronously. Closed-loop CARLA validation across varied weather, lighting, and road-surface conditions shows improved performance over baselines, with complementary friction and visibility adaptations and controller-agnostic integration.","arXiv :2607 .05 180v 1 [ cs .RO] 6 Jul 2026  \nVLM-CASE: vision-language model enabled context-adaptive safety envelopes for anticipatory safe autonomous driving  \nTianjia Yanga , Ke Lib , Ruwen Qinb , Xianbiao (XB) Hua,∗  \na Department of Civil and Environmental Engineering, The Pennsylvania State University, University  \nPark, PA, 16802, United States  \nb Department of Civil Engineering, Stony Brook University, Stony Brook, NY, 11794, United States  \nAbstract  \nAdverse driving conditions, such as bad weather, remain a principal barrier to autonomous driving because they degrade two things at once: what the vehicle can perceive and what it can physically do. Human drivers cope by anticipation, reasoning about the scene and rebudgeting speed, following distance, and steering before grip or sight is lost, whereas current autonomous driving systems at best react after the fact. This paper proposes VLM-CASE, a framework that gives an autonomous vehicle this anticipatory capacity while keeping its motion bounded by a formal safety model at all times. A vision-language model (VLM), fine-tuned with low-rank adaptation (LoRA), reasons about the scene from the front-camera image and reports the road surface and visibility conditions. This output parametrizes a context-adaptive safety envelope (CASE), derived from physical limits and the guarantees of responsibility-sensitive safety, that couples braking and steering through a shared friction budget. A model predictive controller then drives freely within the envelope, while the VLM runs asynchronously so it never blocks the real-time control loop. We validate the framework in closed-loop CARLA simulation on tasks that demand both lateral and longitudinal control, across a range of weather, road-surface, and lighting conditions. The resulting controller, VLM-CASE-MPC, completes all trials, outperforming a conventional MPC baseline anda state-of-the-art VLM-integrated controller. Ablations confirm that the gains come from context adaptation, with the friction and visibility adaptations proving complementary. Furthermore, the framework is controller-agnostic and pairs with almost any low-level controller, offering a promising direction for safe autonomous driving. The dataset and supplementary materials for VLM-CASE are available at [https://github. com/ytj254/VLM-CASE](https://github. com/ytj254/VLM-CASE). Keywords: Vision-language models, Automated vehicles, Responsibility-sensitive safety, Adverse driving conditions, Model predictive control, Safety guarantee  \n1. Introduction  \nHuman drivers handle adverse conditions by anticipation; current autonomous driving systems, at best, by reaction. A driver who notices snow slows before the wheels lose grip,  \n∗ Corresponding author.  \nEmail addresses: [tjyang@psu.edu](tjyang@psu.edu) (Tianjia Yang), [ke.li.1@stonybrook.edu](ke.li.1@stonybrook.edu) (Ke Li),  \n[Ruwen.Qin@stonybrook.edu](Ruwen.Qin@stonybrook.edu) (Ruwen Qin), [xbhu@psu.edu](xbhu@psu.edu) (Xianbiao (XB) Hu)  \nand drops back from the car ahead as fog sets in. Instead of measuring friction and visibility, or consulting a forecast, the driver reasons about the scene ahead and adjusts the driving behavior (speed, following distance, steering) to match what the vehicle can do and what the driver can see. Autonomous driving systems lack this anticipatory capacity, making adverse driving conditions one of the principal barriers to their deployment (Neumeister and Pape, 2019; Boyapati et al., 2023) . Adverse conditions degrade driving in two ways. First, they weaken perception by reducing visibility and impairing lane and object detection (Zang et al. , 2019; Zhang et al., 2023) . Second, they alter vehicle maneuverability by changing tire–road friction and therefore its physical capability (Wang et al., 2022) . Conditions such as rain and snow impair perception and maneuverability at once, and the effects are coupled: the vehicle must make safety-critical decisions precisely when both ","cbCaicbogySJrHDg","https://ap.wps.com/l/cbCaicbogySJrHDg","pdf",10192799,2,1,27,"English","en",105,"# Introduction\n## Motivation: anticipatory safety under adverse conditions\n## Limitations of existing autonomous driving methods\n## Proposed solution overview: VLM-CASE","[{\"question\":\"What problem does VLM-CASE address in autonomous driving under adverse conditions?\",\"answer\":\"It targets the coupled degradation of perception and physical maneuverability caused by bad weather, which current systems often respond to only after the fact rather than anticipating.\"},{\"question\":\"How does VLM-CASE use the vision-language model in the driving pipeline?\",\"answer\":\"A LoRA fine-tuned vision-language model reasons from the front-camera image to report road surface and visibility, which parameterize a context-adaptive safety envelope.\"},{\"question\":\"How does VLM-CASE ensure a formal safety guarantee while adapting to context?\",\"answer\":\"It derives a context-adaptive safety envelope from physical limits and responsibility-sensitive safety, coupling braking and steering through a shared friction budget that constrains an MPC controller at all times.\"}]",1784198306,68,{"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},"vlm-case-vision-language-model-enabled-context-adaptive-safety-envelopes-for-anticipatory-safe-autonomous-driving","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,47,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":20},"https://docshare.wps.com/document/","Document",{"item":48,"name":12,"@type":43,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/vlm-case-vision-language-model-enabled-context-adaptive-safety-envelopes-for-anticipatory-safe-autonomous-driving/84797/",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","2026-07-23","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 problem does VLM-CASE address in autonomous driving under adverse conditions?","Question",{"text":75,"@type":76},"It targets the coupled degradation of perception and physical maneuverability caused by bad weather, which current systems often respond to only after the fact rather than anticipating.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does VLM-CASE use the vision-language model in the driving pipeline?",{"text":80,"@type":76},"A LoRA fine-tuned vision-language model reasons from the front-camera image to report road surface and visibility, which parameterize a context-adaptive safety envelope.",{"name":82,"@type":73,"acceptedAnswer":83},"How does VLM-CASE ensure a formal safety guarantee while adapting to context?",{"text":84,"@type":76},"It derives a context-adaptive safety envelope from physical limits and responsibility-sensitive safety, coupling braking and steering through a shared friction budget that constrains an MPC controller at all times.","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":20,"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"]