To extend the impact of artificial intelligence (AI) on materials and manufacturing, by moving machine intelligence inside manufacturing systems, enabling effective measurement of predictive models…

The National Institute of Standards and Technology awarded a cooperative agreement valued at $398,975.02 to the Massachusetts Institute of Technology on September 1, 2026, under the Measurement and Engineering Research and Standards program (CFDA 11.609).

MIT will develop hardware and software with embedded sensors, actuators, and processors to enable artificial intelligence application to manufacturing processes. The effort combines AI, materials research and development, and manufacturing metrology to move machine intelligence inside manufacturing systems, measuring predictive models rather than single parameters. Researchers will develop and implement software models of materials, feedback controls, distributed communications, and data management. Initial instrumentation and development will focus on machining, with additional foci including extrusion, cutting, and reacting. Instrumentation includes high-speed multi-axis MEMS IMUs for vibration measurement, multi-axis flexural load cells for mechanical load measurement, current mirror H-bridges for motor load measurement, self-calibrating vision systems for motion measurement, imaging interferometers for displacement measurement, and spectrometers for excitation measurement. Development proceeds in three phases: external, out-of-band manufacturing machine learning components; in-band system integration components; and native manufacturing machine learning systems. The effort will produce machines, methods, software, and associated protocols for incorporating AI into materials manufacturing and demonstrate how to embed machine learning into instrumented manufacturing processes.

Performance occurs in Cambridge, Massachusetts, with a period of performance from September 1, 2026, through August 31, 2029.

Activities

The researchers will develop hardware and software with incorporated sensors, actuators and processors to enable the application of AI to manufacturing, supported by the development and implementation of software models of materials, feedback controls, communication, and data management. Initial instrumentation and development of the paradigm will focus on machining, with additional foci including extrusion, cutting and reacting. Instrumentation will include:-high-speed multi-axis mems imus for measuring vibrations-multi-axis flexural load cells for measuring mechanical loads-current mirror h-bridges for measuring motor loads-self-calibrating vision systems for measuring motion-imaging interferometers for measuring displacements-spectrometers for measuring excitationsspecific plans include 1) development and testing of the manufacturing machine learning (MML) components that can be installed externally for out-of-band applications; 2) developing and testing MML components for in-band system integration and finally; (3) developing and testing native MML systemsexpected outcomes: this effort will result in a set of machines, methods, software, and associated protocols for realizing the incorporation of AI into the manufacturing of materials. The effort will demonstrate how to embed machine learning into instrumented manufacturing processes in order to improve their speed, security, stability, scalability, portability, and adaptability.

Intended beneficiaries

The beneficiaries of this effort will include MIT and NIST, who will be collaborators on the project, and thereby acquire these methods and models, and, ultimately, the general public, who will see, through the dissemination of these methods and models, these approaches translated into better performing materials and manufacturing products at lower cost.

Subrecipient activities

The recipient does not intend to subaward funds.