The Pulse
DOE Selects Four Lab Robotics Projects in a $30 Million Program
The Department of Energy selected four national laboratory-led robotics projects on October 8, 2026, under a program with $30 million in planned funding. The projects target reusable lab-robot skills, reliability testing, connected experime
AI.info Team · Written with AI ·
Four labs, four links in an autonomous-science chain
The U.S. Department of Energy selected four national laboratory-led robotics projects on October 8, setting out a $30 million program aimed at making autonomous scientific work transferable across research facilities. The selections span systems that learn laboratory tasks, methods for testing whether robots perform them reliably, software to connect experiments and computing, and automation at photon- and electron-science facilities.
The projects are led by Argonne, Brookhaven, Oak Ridge and SLAC national laboratories. DOE expects them to produce open software and interfaces, reusable robot skills, digital-twin environments, benchmark tasks, datasets, trained models, safety practices, provenance records and training resources. The Office of Science put the aim this way:
“The objective is not simply to produce four impressive robotics demonstrations, but to establish capabilities that can be reused and extended by the broader DOE scientific community.”
U.S. Department of Energy Office of Science, selection announcement, October 8, 2026
Argonne builds skills; Brookhaven tries to break them
Argonne’s Modular Autonomous Experimentation through Self-improving Testbeds for Robotic Operations, or MAESTRO, will combine digital twins, world models, reusable skills, agent-based orchestration and safety mechanisms. Its proposed robotics sandbox and science-focused proving grounds are meant to test whether robots can improve through operational experience and carry skills between instruments and laboratory workflows.
Brookhaven’s DOE AI Robotics Testbed to Generalize Autonomous Science, or DART, takes on the reliability question. The project plans adversarial digital counterparts, a verified failure atlas and methods for transferring knowledge between robot platforms. Its evaluations will look beyond whether a robot completes a movement: DOE says the team will assess whether the physical result and scientific outcome satisfy the experiment’s requirements.
Oak Ridge connects the tools; SLAC targets its light and electron sources
Oak Ridge’s Testbed for Robotics and Autonomy in Connected Experiments, or TRACE, will build on the lab’s INTERSECT automated-laboratory ecosystem. It aims to connect robots, scientific instruments, what DOE calls “SI agents” (the department’s abbreviation for Super Intelligence), digital twins, data systems and computing through standardized interfaces. The project also plans to record the chain of experimental decisions and results, test systems under different levels of simulation fidelity, and make evaluations independently repeatable.
SLAC’s SPIRE project focuses on photon- and electron-science facilities. It will combine physical sample handling with intelligent detectors, accelerator and instrument controls, persistent experiment state, and computing from the edge through high-performance systems. DOE says the facilities require decisions on timescales ranging from microseconds to hours; SPIRE is intended to develop methods and benchmarks that other user facilities can adapt.
The selections are not yet final awards
The projects sit under the Office of Science’s Advanced Scientific Computing Research program and its LAB 26-3601 solicitation. That call sought reusable, interoperable testbeds that combine robotics with AI, scientific instruments and advanced computing, with evaluation measures such as reproducibility, provenance and the need for human intervention. It initially projected two to three awards, each expected to run for three years with an estimated $3 million to $5 million per year; DOE’s October announcement names four selections and does not specify each project’s share.
DOE lists $30 million in total planned funding, including $2 million for fiscal year 2026, with later-year funding dependent on congressional appropriations. The department also cautions that selection starts award negotiations rather than guaranteeing an award: it can cancel negotiations or rescind a selection before funding is issued.