{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T04:46:35Z","timestamp":1785732395713,"version":"3.56.0"},"reference-count":30,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T00:00:00Z","timestamp":1785715200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Robot. AI"],"abstract":"<jats:p>Robots that execute language-conditioned tasks in dynamic environments often rely on feedback only after an action has failed, which can be insufficient when failures involve collisions or workspace conflicts. This paper presents a predictive monitoring framework that uses Vision-Language Models (VLMs) to assess near-future execution risk during robot task execution. The framework first generates structured plans with action execution conditions and a plan-level fallback action. During execution, a monitoring module combines visual observations, the current action, and the relevant execution conditions to estimate whether a condition is likely to be violated within a short future time window. When the predicted risk exceeds a task-specific threshold, the robot halts the current action, executes the fallback behavior, and replans from the updated state. We evaluate the approach in Gazebo simulation on mobile navigation with a moving human obstacle and manipulation with two robot arms sharing a workspace. Across controlled collision-risk settings, the proposed method achieves higher task success rates than reactive VLM-based baselines while requiring fewer replanning events than conservative current-state precondition checking. The results indicate that predictive vision-language monitoring can improve task completion in simulated dynamic robot tasks, while remaining subject to limitations such as VLM latency, prompt sensitivity, and evaluation beyond simulation.<\/jats:p>","DOI":"10.3389\/frobt.2026.1870024","type":"journal-article","created":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T04:28:09Z","timestamp":1785731289000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Predictive vision-language monitoring for proactive safety in robot task execution"],"prefix":"10.3389","volume":"13","author":[{"given":"Juan Diego","family":"Pe\u00f1a-Narvaez","sequence":"first","affiliation":[{"name":"Intelligent Robotics Lab, Signal Theory, Communications, Telematics Systems, and Computation Department, International Doctoral School, Rey Juan Carlos University","place":["Fuenlabrada, Spain"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rodrigo","family":"P\u00e9rez-Rodr\u00edguez","sequence":"additional","affiliation":[{"name":"Intelligent Robotics Lab, Universidad Rey Juan Carlos","place":["Fuenlabrada, Spain"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Esther","family":"Aguado","sequence":"additional","affiliation":[{"name":"Intelligent Robotics Lab, Universidad Rey Juan Carlos","place":["Fuenlabrada, Spain"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Francisco Miguel","family":"Moreno","sequence":"additional","affiliation":[{"name":"Intelligent Robotics Lab, Universidad Rey Juan Carlos","place":["Fuenlabrada, Spain"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Francisco Mart\u00edn","family":"Rico","sequence":"additional","affiliation":[{"name":"Intelligent Robotics Lab, Universidad Rey Juan Carlos","place":["Fuenlabrada, Spain"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,8,3]]},"reference":[{"key":"B1","first-page":"287","article-title":"Do as i can, not as i say: grounding language in robotic affordances","volume-title":"Conference on Robot Learning (Corl)","author":"Ahn","year":"2022"},{"key":"B2","doi-asserted-by":"publisher","first-page":"4340","DOI":"10.1109\/ICRA57147.2024.10610784","article-title":"How to prompt your robot: a promptbook for manipulation skills with code as policies","author":"Arenas","year":"2024","journal-title":"2024 IEEE International Conference on Robotics and Automation (ICRA)"},{"key":"B3","volume-title":"Commonsense Reasoning for Legged Robot Adaptation with vision-language Models","author":"Chen","year":"2024"},{"key":"B4","volume-title":"Planning with Reasoning Using Vision Language World Model","author":"Chen","year":"2025"},{"key":"B5","doi-asserted-by":"publisher","first-page":"105303","DOI":"10.1016\/j.robot.2025.105303","article-title":"Cresta: a cognitivist robot execution framework for semantic-driven task awareness","volume":"197","author":"Gasperini","year":"2025","journal-title":"Robotics Aut. Syst."},{"key":"B6","first-page":"12124","volume-title":"Doremi: Grounding Language Model by Detecting and Recovering from plan-execution Misalignment","author":"Guo","year":"2024"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2311.17842","author":"Hu","year":"2023","journal-title":"Look Before you Leap: Unveiling the Power of gpt-4v in Robotic vision-language PlanningarXiv Preprint arXiv:2311"},{"key":"B8","first-page":"1769","article-title":"Inner monologue: embodied reasoning through planning with language models","volume-title":"Conference on Robot Learning (Corl)","author":"Huang","year":"2022"},{"key":"B9","doi-asserted-by":"crossref","first-page":"4412","DOI":"10.18653\/v1\/2020.findings-emnlp.395","article-title":"Visually-grounded planning without vision: language models infer detailed plans from high-level instructions","volume-title":"Findings of the Association for Computational Linguistics: Emnlp 2020","author":"Jansen","year":"2020"},{"key":"B10","doi-asserted-by":"publisher","first-page":"4857","DOI":"10.1109\/ICRA57147.2024.10610434","article-title":"Copal: corrective planning of robot actions with large language models","author":"Joublin","year":"2024","journal-title":"2024 IEEE International Conference on Robotics and Automation (ICRA)"},{"key":"B11","doi-asserted-by":"publisher","first-page":"1091","DOI":"10.1007\/s11370-024-00550-5","article-title":"A survey on integration of large language models with intelligent robots","volume":"17","author":"Kim","year":"2024","journal-title":"Intell. Serv. Robot."},{"key":"B12","first-page":"2679","article-title":"Openvla: an open-source vision-language-action model","volume-title":"Conference on Robot Learning","author":"Kim","year":"2025"},{"key":"B13","doi-asserted-by":"publisher","first-page":"6872","DOI":"10.1109\/lra.2024.3412593","article-title":"Fine-grained task planning for service robots based on object ontology knowledge via large language models","volume":"9","author":"Li","year":"2024","journal-title":"IEEE Robotics Automation Lett. (RA-L)"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2304.11477","author":"Liu","year":"","journal-title":"Llm+P: Empowering Large Language Models with Optimal Planning ProficiencyarXiv Preprint arXiv:2304"},{"key":"B15","first-page":"3468","article-title":"Reflect: summarizing robot experiences for failure explanation and correction","volume-title":"Conference on Robot Learning (Corl)","author":"Liu","year":""},{"key":"B16","article-title":"Seeing before observable: potential risk reasoning in autonomous driving via vision language models","author":"Liu","year":"2025"},{"key":"B17","first-page":"6273","volume-title":"Skiros2: A Skill-based Robot Control Platform for Ros","author":"Mayr","year":"2023"},{"key":"B18","doi-asserted-by":"publisher","first-page":"10201","DOI":"10.1109\/lra.2024.3471457","article-title":"Replanvlm: replanning robotic tasks with visual language models","volume":"9","author":"Mei","year":"2024","journal-title":"IEEE Robotics Automation Lett. (RA-L)"},{"key":"B19","first-page":"11523","article-title":"Progprompt: generating situated robot task plans using large language models","volume-title":"IEEE International Conference on Robotics and Automation (ICRA)","author":"Singh","year":"2023"},{"key":"B20","doi-asserted-by":"publisher","first-page":"2998","DOI":"10.1109\/ICCV51070.2023.00280","article-title":"Llm-planner: Few-shot grounded planning for embodied agents with large language models","author":"Song","year":"2023","journal-title":"2023 IEEE\/CVF International Conference on Computer Vision (ICCV)"},{"key":"B21","first-page":"66383","article-title":"Chain-of-thought reasoning without prompting","volume-title":"Advances in Neural Information Processing SystemsAdvances in Neural Information Processing Systems","author":"Wang","year":"2024"},{"key":"B22","doi-asserted-by":"publisher","first-page":"24824","DOI":"10.52202\/068431-1800","article-title":"Chain-of-thought prompting elicits reasoning in large language models","volume":"35","author":"Wei","year":"2022","journal-title":"Adv. Neural Information Processing Systems"},{"key":"B23","doi-asserted-by":"publisher","first-page":"4003","DOI":"10.1109\/lra.2024.3371223","article-title":"Text2reaction: enabling reactive task planning using large language models","volume":"9","author":"Yang","year":"2024","journal-title":"IEEE Robotics Automation Lett. (RA-L)"},{"key":"B24","volume-title":"Statler: State-Maintaining Language Models for Embodied Reasoning","author":"Yoneda","year":"2023"},{"key":"B25","article-title":"Safevla: towards safety alignment of vision-language-action model via constrained learning","author":"[Dataset] Zhang","year":"2025"},{"key":"B26","doi-asserted-by":"publisher","first-page":"31967","DOI":"10.52202\/075280-1387","article-title":"Large language models as commonsense knowledge for large-scale task planning","volume":"36","author":"Zhao","year":"2023","journal-title":"Adv. Neural Information Processing Systems"},{"key":"B27","doi-asserted-by":"publisher","first-page":"2942","DOI":"10.1109\/LRA.2025.3533476","article-title":"A dual-agent collaboration framework based on llms for nursing robots to perform bimanual coordination tasks","volume":"10","author":"Zhao","year":"2025","journal-title":"IEEE Robotics Automation Lett."},{"key":"B28","doi-asserted-by":"publisher","first-page":"2081","DOI":"10.1109\/ICRA57147.2024.10610065","article-title":"Isr-llm: iterative self-refined large language model for long-horizon sequential task planning","author":"Zhou","year":"2024","journal-title":"2024 IEEE International Conference on Robotics and Automation (ICRA)"},{"key":"B29","first-page":"6919","article-title":"Code-as-monitor: Constraint-aware visual programming for reactive and proactive robotic failure detection","volume-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Zhou","year":"2025"},{"key":"B30","first-page":"2165","article-title":"Rt-2: Vision-language-action models transfer web knowledge to robotic control","volume-title":"Conference on Robot Learning","author":"Zitkovich","year":"2023"}],"container-title":["Frontiers in Robotics and AI"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frobt.2026.1870024\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T04:28:12Z","timestamp":1785731292000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frobt.2026.1870024\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,3]]},"references-count":30,"alternative-id":["10.3389\/frobt.2026.1870024"],"URL":"https:\/\/doi.org\/10.3389\/frobt.2026.1870024","relation":{},"ISSN":["2296-9144"],"issn-type":[{"value":"2296-9144","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,3]]},"article-number":"1870024"}}