{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T21:20:14Z","timestamp":1792790414859,"version":"4.5.0"},"reference-count":22,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,7,17]],"date-time":"2025-07-17T00:00:00Z","timestamp":1752710400000},"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:sec><jats:title>Introduction<\/jats:title><jats:p>Mission-critical automation demands decision-making that is explainable, adaptive, and scalable\u2014attributes elusive to purely symbolic or data-driven approaches. We introduce a hybrid intelligence (H-I) system that fuses symbolic reasoning with advanced machine learning <jats:italic>via<\/jats:italic> a hierarchical architecture, inspired by cognitive frameworks like Global Workspace Theory (Baars, A Cognitive Theory of Consciousness, 1988).<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>This architecture operates across three levels to achieve autonomous, end-to-end workflows: Navigation: Using Vision Transformers, and graph-based neural networks, the system navigates file systems, databases, and software interfaces with precision. Discrete Actions: Multi-framework automated machine learning (AutoML) trains agents to execute discrete decisions, augmented by Transformers and Joint Embedding Predictive Architectures (JEPA) (Assran et al., 2023, 15619\u201315629) for complex time-series analysis, such as anomaly detection. Planning: Reinforcement learning, world model-based reinforcement learning, and model predictive control orchestrate adaptive workflows tailored to user requests or live system demands.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>The system\u2019s capabilities are demonstrated in two mission-critical applications: Space Domain Awareness, Satellite Behavior Detection: A graph-based JEPA paired with multi-agent reinforcement learning enables near real-time anomaly detection across 15,000 on-orbit objects, delivering a precision-recall score of 0.98. Autonomously Driven Simulation Setup: The system autonomously configures Computational Fluid Dynamics (CFD) setups, with an AutoML-driven optimizer enhancing the meshing step\u2014boosting boundary layer capture propagation (BL-CP) from 8% to 98% and cutting geometry failure rates from 88% to 2% on novel aircraft geometries. Scalability is a cornerstone, with the distributed training pipeline achieving linear scaling across 2,000 compute nodes for AI model training, while secure model aggregation incurs less than 4% latency in cross-domain settings.<\/jats:p><\/jats:sec><jats:sec><jats:title>Discussion<\/jats:title><jats:p>By blending symbolic precision with data-driven adaptability, this hybrid intelligence system offers a robust, transferable framework for automating complex knowledge work in domains like space operations and engineering simulations\u2014and adjacent applications such as autonomous energy and industrial facility operations, paving the way for next-generation industrial AI systems.<\/jats:p><\/jats:sec>","DOI":"10.3389\/frobt.2025.1566623","type":"journal-article","created":{"date-parts":[[2025,7,17]],"date-time":"2025-07-17T04:10:38Z","timestamp":1752725438000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":6,"title":["Hybrid intelligence systems for reliable automation: advancing knowledge work and autonomous operations with scalable AI architectures"],"prefix":"10.3389","volume":"12","author":[{"given":"Allan","family":"Grosvenor","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anton","family":"Zemlyansky","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Abdul","family":"Wahab","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kyrylo","family":"Bohachov","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aras","family":"Dogan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dwyer","family":"Deighan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2025,7,17]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"15619","DOI":"10.1109\/cvpr52729.2023.01499","article-title":"Self-supervised learning from images with a joint-embedding predictive architecture","author":"Assran","year":"2023","journal-title":"arXiv"},{"key":"B2","volume-title":"A cognitive theory of consciousness","author":"Baars","year":"1988"},{"key":"B3","doi-asserted-by":"publisher","first-page":"42","DOI":"10.1109\/HiPC.2018.00014","article-title":"DeepHyper: asynchronous hyper-parameter search for deep neural networks","author":"Balaprakash","year":"2018"},{"key":"B4","doi-asserted-by":"publisher","first-page":"024001","DOI":"10.1088\/1361-6471\/abd009","article-title":"Optimization and supervised machine learning methods for fitting numerical physics models without derivatives","volume":"48","author":"Bollapragada","year":"2020","journal-title":"J. Phys. G."},{"key":"B5","unstructured":"CelesTrak two-line element data"},{"key":"B6","article-title":"Decision transformer: reinforcement learning via sequence modeling","author":"Chen","year":"2021"},{"key":"B7","doi-asserted-by":"crossref","first-page":"785","DOI":"10.1145\/2939672.2939785","article-title":"XGBoost: a scalable tree boosting system","volume-title":"Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","author":"Chen","year":"2016"},{"key":"B8","first-page":"1","article-title":"Optimizing distributed training on frontier for large language models","volume-title":"ISC high performance","author":"Dash","year":"2024"},{"key":"B9","first-page":"337","article-title":"Z3: an efficient SMT solver","author":"de Moura","year":"2008"},{"key":"B10","article-title":"Machine learning for parameter auto-tuning in molecular-dynamics simulations","author":"Kadupitiya","year":"2019"},{"key":"B11","first-page":"12086","article-title":"BLIP: bootstrapping language-image pre-training for unified vision-language understanding and generation","author":"Li","year":"2022"},{"key":"B12","doi-asserted-by":"publisher","first-page":"239","DOI":"10.1080\/00401706.1979.10489755","article-title":"Comparison of three methods for selecting values of input variables in the analysis of output from a computer code","volume":"21","author":"McKay","year":"1979","journal-title":"Technometrics"},{"key":"B13","volume-title":"The society of mind","author":"Minsky","year":"1986"},{"key":"B14","article-title":"MSBAI unveils revolutionary AI copilot for space traffic control at CES 2025 [Press release]","year":"2025"},{"key":"B15","article-title":"A time series is worth 64 words: long-term forecasting with transformers","author":"Nie","year":"2022"},{"key":"B16","article-title":"PyTorch: an imperative style, high-performance deep learning library","author":"Paszke","year":"2019"},{"key":"B17","article-title":"Learning transferable visual models from natural language supervision","author":"Radford","year":"2021"},{"key":"B18","article-title":"ZeRO: memory optimizations toward training trillion-parameter models","author":"Rajbhandari","year":"2019"},{"key":"B19","article-title":"Proximal policy optimization algorithms","author":"Schulman","year":"2017"},{"key":"B20","doi-asserted-by":"publisher","first-page":"433","DOI":"10.1016\/j.concog.2005.11.005","article-title":"A cognitive architecture that combines global workspace and predictive processing","volume":"15","author":"Shanahan","year":"2020","journal-title":"Conscious. Cogn."},{"key":"B21","article-title":"Graph-level representation learning with joint-embedding predictive architectures","author":"Skenderi","year":"2025"},{"key":"B22","unstructured":"United States Space Force, Colorado Springs, CO, USA"}],"container-title":["Frontiers in Robotics and AI"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frobt.2025.1566623\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,17]],"date-time":"2025-07-17T04:10:40Z","timestamp":1752725440000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frobt.2025.1566623\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,17]]},"references-count":22,"alternative-id":["10.3389\/frobt.2025.1566623"],"URL":"https:\/\/doi.org\/10.3389\/frobt.2025.1566623","relation":{},"ISSN":["2296-9144"],"issn-type":[{"value":"2296-9144","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,17]]},"article-number":"1566623"}}