Vision-Guided Robotic Automation and Digital Twin Framework for Autonomous Experiments

Brookhaven National Laboratory, operating under the Department of Energy Office of Science, intends to solicit a cost-reimbursement research subcontract for vision-guided robotic automation and digital twin framework development for autonomous experiments, with solicitation RFQ-447848 anticipated to release in or around September 1, 2026. This procurement is not set-aside and will be competed full and open.

Work is performed at Brookhaven National Laboratory on Long Island, New York. The scope encompasses three primary technical components: deployment of stereo camera systems with 3D perception pipelines and deep-learning-based object detection for sample holders, detectors, robotic tools, and environmental fixtures in laboratory and beamline environments; development of a digital twin synchronizing real-time perception data with robotic kinematics, workspace geometry, beamline components, and mock hardware; and a robotic simulation environment incorporating kinematic models and perception data to perform pre-execution reachability analysis, collision prediction, clearance margin estimation, and singularity detection. The Base Year spans 12 months, with hardware setup and deep-learning model training in months 0–2, commissioning for QAS beamline and laboratory in months 3–5, TES beamline commissioning in months 6–9, and algorithm optimization with paper preparation in months 9–12. Optional continuation work for Year 1 and Year 2 includes expanded system deployment, advanced deep-learning optimization, digital twin geometry development, and validation of simulation predictions against experimental results. A single award is anticipated. Interested parties must be registered under NAICS 611310 (Colleges, Universities, and Professional Schools) in SAM prior to submission. The draft Statement of Work is attached. The response deadline is July 13, 2026.