{"id":64625,"date":"2026-07-24T19:47:20","date_gmt":"2026-07-24T23:47:20","guid":{"rendered":"https:\/\/overcentral.com\/en\/?p=64625"},"modified":"2026-07-24T19:47:20","modified_gmt":"2026-07-24T23:47:20","slug":"mit-autonomous-nuclear-control-system","status":"publish","type":"post","link":"https:\/\/overcentral.com\/en\/mit-autonomous-nuclear-control-system\/","title":{"rendered":"MIT Develops Autonomous Control System for Nuclear Plants"},"content":{"rendered":"

For decades, the nuclear power industry has operated on a fundamental assumption: that large, centralized plants running at full capacity can justify the cost of maintaining a large, highly skilled workforce. But as the industry pivots toward smaller, distributed microreactors\u2014particularly those intended for remote and rural locations\u2014that economic equation breaks down. The solution, emerging from the Massachusetts Institute of Technology<\/a>, is a new breed of autonomous control system designed not to replace human operators entirely, but to orchestrate a fluid, intelligent partnership between humans and machines. This is not a theoretical exercise. MIT doctoral researcher Lauren Fortier, building on her master’s thesis completed in 2025, is developing a supervisory control system that could fundamentally alter how nuclear plants are operated, managed, and staffed, with profound implications for the future of clean energy<\/a> deployment.<\/p>\n

Why Autonomous Control Matters for the Next Generation of Nuclear Plants<\/h2>\n

Legacy nuclear power plants are engineering marvels of manual operation. They require large teams of operators, engineers, and technicians to manage complex procedures, safety checks, and routine maintenance. That model works when a plant generates hundreds of megawatts and runs at baseload capacity. But the future of nuclear energy, as envisioned by many industry leaders and research institutions, includes fleets of small modular reactors and microreactors deployed in remote areas\u2014mining sites, isolated communities, industrial facilities, or disaster zones. In such settings, the economics of maintaining a full-time human operations staff become prohibitive. The question Fortier and her collaborators at MIT, the Idaho National Laboratory<\/a> (INL), and Westinghouse<\/a> are tackling is not simply whether automation can work, but how to design a system that is rigorous, transparent, and trustworthy enough to handle the safety-critical demands of nuclear power.<\/p>\n

The Core Problem: Human-Centric Procedures Don’t Fit Machines<\/h3>\n

Fortier’s research began with a fundamental observation: every existing procedure in a nuclear plant is designed for human execution. Humans read instructions, turn valves, check gauges, and follow step-by-step protocols. When you try to hand those same procedures to a computer, the rigidity becomes a liability. “Because everything is human-centric, it doesn’t allow you to choose the best way to do a procedure,” Fortier notes. Her insight was that a truly effective autonomous control system could not simply automate existing human workflows. It needed to create a new framework where humans and computers could each do what they do best, with strategic human intervention delivered only when necessary. This is the philosophical foundation of what she calls a “supervisory control system”\u2014a single, integrated approach rather than a patchwork of interlinked parts.<\/p>\n

How MIT and Idaho National Laboratory Are Collaborating on Human-Machine Interfaces<\/h2>\n

Fortier’s path to this solution was shaped by cross-disciplinary collaborations that are a hallmark of MIT’s approach. Her research advisor, Sacit Cetiner, holds a joint appointment with MIT’s Department of Nuclear Science and Engineering (NSE) and the Idaho National Laboratory. That connection gave Fortier direct access to INL’s Human System Simulation Laboratory, where she worked with senior human factors scientist Katya Le Blanc. The collaboration was essential because Fortier, as she freely admits, is “very much an engineer and don’t have a lot of experience in human behavior.” The human factors expertise helped her understand what operators need to see and do when a machine fails or requires human override. “I got better insights into many aspects, including what you want to see when a human has to take over for a machine when it’s no longer working,” she says.<\/p>\n

This human-machine interface challenge is not a peripheral concern. In any autonomous system, the moments when control shifts from computer to human are the most vulnerable. The design must make that transition intuitive, clear, and error-resistant. INL’s experience in simulating control room environments gave Fortier a practical testing ground for her ideas.<\/p>\n

Westinghouse and the Industry Testing Ground<\/h3>\n

Fortier also spent the summer of 2025 as an intern at Westinghouse, a leading vendor for current and next-generation nuclear plants. That experience allowed her to test her concepts against real-world constraints and industry expectations. The combination of academic control theory, national laboratory human factors research, and industry vendor perspective created a rare trifecta of expertise feeding into the system’s design.<\/p>\n

The Control Theory Foundation: Learning from MIT’s Active-Adaptive Control Laboratory<\/h2>\n

One of Fortier’s co-advisors is Anuradha Annaswamy, a founder and director of the Active-Adaptive Control Laboratory in MIT’s Department of Mechanical Engineering. Annaswamy is a control systems expert, and she provides the mathematical and theoretical scaffolding for Fortier’s supervisory control framework. “She’s a control systems expert, which really benefits me because while I can explain what to do with a nuclear power plant, she can help me understand better how to go about operations from a control systems perspective,” Fortier explains. Fortier has taken graduate-level control theory classes to build a foundation in the discipline, ensuring that her automation framework is grounded in rigorous engineering principles rather than ad-hoc heuristics. Her other co-advisor is Curtis Smith, the former director for INL’s Nuclear Safety and Regulatory Research Division, now a KEPCO Professor of the Practice of Nuclear Science and Engineering at MIT NSE.<\/p>\n

What Is Finite State Automata and Why It Matters for Nuclear Safety<\/h2>\n

One of the most critical decisions in Fortier’s research is the choice of automation technology. She is not using machine learning or artificial intelligence. Instead, her system is built on a technique called finite state automata (FSA). This is a discrete event system where every action is event-driven: if a specific condition occurs, the system executes a specific response. Every state transition is clearly defined and transparent. Unlike a neural network, where the internal logic can be opaque, FSA allows operators and regulators to trace<\/a> exactly why the system made a particular decision. “We’re not using a data-driven statistical approach like machine learning because we do not yet have the tools to validate the operation of such systems,” Fortier states. This is a crucial point for nuclear regulation, where the ability to verify and validate every aspect of a control system’s behavior is non-negotiable. Fortier studied FSA extensively during an internship at INL in the summer of 2024, and it has become the backbone of her approach.<\/p>\n

A Step-by-Step Path to Build Trust: Gradual Autonomy and Objective-Oriented Operations<\/h2>\n

Fortier is acutely aware that introducing automation into a nuclear plant requires building trust with operators, regulators, and the public. Her system is designed to progress incrementally. “When we introduce an automated procedure that walks you step by step through what you would be doing anyway, it is reassuring and builds trust,” she notes. The first phase involves automating existing procedures in a way that feels familiar to operators. The next phase, which is the focus of her doctoral work, is more ambitious: objective-oriented operations. In this model, the control system itself creates the sequence of events needed to reach a given objective, rather than following a predetermined, static procedure. That means the system can adapt to current plant conditions, adjusting its actions in real time. This is a significant departure from the rigid, step-by-step procedures that dominate current nuclear operations.<\/p>\n

DOE Recognition: Fortier Wins 2025 Innovations in Nuclear Energy Competition<\/h2>\n

The promise of this work has already received official recognition. Fortier was one of the winners of the 2025 edition of the Innovations<\/a> in Nuclear Energy Research and Development Student Competition from the U.S. Department of Energy’s Nuclear Energy University Program (NEUP). That award underscores the government’s interest in advancing autonomous control technologies as a key enabler for commercial microreactor deployment.<\/p>\n

Scaling the Supervisory Control System: From Theory to Commercial Microreactors<\/h2>\n

The immediate next step for Fortier is scaling her supervisory control system. She has built a proof-of-concept for a small aspect of control, and now the challenge is to expand it to cover the full scope of plant operations. The work is directly relevant to the development of commercial microreactors, which cannot support the large staffing footprint of legacy plants. The collaborations Fortier has built\u2014with INL, Westinghouse, and MIT’s control theory experts\u2014have given her a real-world testing ground and industry relevance. “The collaborations with other people, and the relationships we have established with stakeholders, have really helped make an impact and supported the relevancy of the work,” she says. “Sometimes when you’re stuck in your own bubble, that outside perspective is really useful.”<\/p>\n

As the nuclear industry looks to deploy small, clean, and reliable power sources in places where traditional plants cannot go, the work being done at MIT today may provide the control backbone that makes that vision a reality. The path is cautious, methodical, and grounded in proven engineering principles\u2014exactly what the safety-critical nuclear domain demands. Fortier’s autonomous control system is not a bid to eliminate human operators, but to enable a future where nuclear power can go where it is needed most, even when the people to run it are far away.<\/p>\n","protected":false},"excerpt":{"rendered":"

For decades, the nuclear power industry has operated on a fundamental assumption: that large, centralized plants running at full capacity can justify the cost of maintaining a large, highly skilled workforce. But as the industry pivots toward smaller, distributed microreactors\u2014particularly those intended for remote and rural locations\u2014that economic equation breaks down. The solution, emerging from […]<\/p>\n","protected":false},"author":7,"featured_media":83761,"comment_status":"closed","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"fifu_image_url":"https:\/\/cards.overcentral.com\/cards\/en\/64625.png","fifu_image_alt":"MIT Develops Autonomous Control System for Nuclear Plants","footnotes":""},"categories":[349],"tags":[],"class_list":["post-64625","post","type-post","status-publish","format-standard","has-post-thumbnail","category-articles"],"fifu_image_url":"https:\/\/cards.overcentral.com\/cards\/en\/64625.png","fifu_image_alt":"MIT Develops Autonomous Control System for Nuclear Plants","_links":{"self":[{"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/posts\/64625","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/users\/7"}],"replies":[{"embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/comments?post=64625"}],"version-history":[{"count":0,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/posts\/64625\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/media\/83761"}],"wp:attachment":[{"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/media?parent=64625"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/categories?post=64625"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/tags?post=64625"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}