AI-driven prosthetic limbs that restore movement and sensation — closing the loop between nerve, machine, and touch.
Utah Neurorobotics Lab
venture active confidence: Medium status: Draft updated 2026-07-14
Summary
The Utah Neurorobotics Lab is a University of Utah engineering group led by Jacob George (official site). Its bidirectional interfaces join Utah's electrode-array legacy to modern decoding and sensory feedback—a credible bridge from machine learning to restored function.
Impact
Upper-limb prosthetics have been stuck for decades on myoelectric control from surface EMG, typically one or two degrees of freedom with no sensory feedback. Real hand dexterity requires many degrees of freedom and fine tactile discrimination. The George lab's bidirectional framing is specific: existing prosthetics are open-loop (decode intent → actuate); bidirectional interfaces close the loop (decode → actuate → sense → feedback).
Patients in George lab studies have reported "feeling" their prosthetic hand — their brains processing stimulation signals as touch. The decoding problem is also a frontier applied-ML challenge: peripheral nerve signals are noisy, non-stationary, and multi-modal. Impact horizons are measured in years to decades, with translation likely through startup spinouts leveraging Utah Array–class interface technology.
What They Are Building
Two linked technical problems anchor the lab:
- Decode — translate ongoing peripheral nerve signals into intended joint trajectories with low latency, high accuracy, and tolerance to electrode drift and day-to-day variation.
- Encode — deliver electrical stimulation patterns that evoke graded, localized, naturalistic tactile sensations rather than painful buzzes.
The work builds on the Utah Slanted Electrode Array (USEA), a landmark neural-interface technology developed at the University of Utah. Modern transformer and state-space architectures are increasingly applied to the decoding side. Real-world deployment adds months of continuous wear, sweat, movement artifacts, and the brain's adaptive reorganization — all of which make both problems harder outside the lab. Official Website: Utah Neurorobotics Lab
What They Need Now
The lab recruits graduate students, postdocs, and research engineers in machine learning, biomedical engineering, electrical engineering, and clinical-research support. Funding is grant-driven (NIH, NSF, DoD including DARPA). IP flows through the University of Utah Technology Licensing Office and may seed startups over a 5–10 year horizon.
For talent matching, this fits people who want academic research culture — publication-oriented, long timelines — at the intersection of ML, neuroscience, and clinical impact.
Who Could Help
Useful helpers include neural-interface commercialization advisors, FDA and clinical-trial consultants for eventual device translation, Utah Technology Licensing Office staff, and partners in the Utah Array ecosystem (e.g., Blackrock Neurotech and related spinout pipelines). Grant-writing and DoD program managers familiar with neural-engineering portfolios are relevant.
Utah Context
The lab continues a deep University of Utah tradition in neural engineering and connects Utah's machine-intelligence talent pool to health and longevity applications. It is an academic lab, not a company — but it is plausibly among the stronger single ML-applied-to-neural-interfaces programs in the state and deserves visibility for anyone considering PhD or research-staff roles in neural engineering.
Evidence
- Official Website: Utah Neurorobotics Lab · https://neurorobotics.utah.edu
- Official Website: Utah NeuroRobotics Lab (ECE) · https://neurorobotics.ece.utah.edu
Open Questions
- What are the lab's current enrollment, funding totals, and nearest clinical-translation milestones?
- Which spinout candidates, if any, are in the U of U licensing pipeline?
- Primary lab website was intermittently unavailable during sourcing; claims should be re-verified against publications and NIH RePORTER.