Plan4Act · Project

Brain-Controlled Smart Home System Letting Disabled People Plan Actions by Thought

healthPrototypeTRL 4Thin data (2/5)

Imagine thinking "I want coffee" and your smart home starts making it — without you having to mentally command every single step like "move arm, grab cup, press button." Right now, brain-controlled devices work like a TV remote: one button press, one action. This project figured out how to read the brain's planning signals — the part where you think ahead about a whole sequence — and translate that into a chain of smart home commands. They built a working hardware controller that connects brain signals to home devices, tested with real neural data from primates.

By the numbers

EUR 4,236,000

EU contribution funding the research

TRL4

Technology readiness level achieved

Consortium partners

Countries involved (DE, DK, ES)

Total project deliverables produced

Demonstration deliverables (software and hardware demos)

The business problem

What needed solving

Current brain-machine interfaces are reactive — they translate one thought into one command, forcing motor-impaired users to mentally trigger every single step of a multi-step task. This makes even simple daily activities like making coffee exhausting and impractical. There is no commercial system that can read a person's planned action sequence from brain signals and execute it proactively across connected home devices.

The solution

What was built

The project built an integrated system at TRL4 consisting of: neural signal recording and decoding methods for predictive brain activity, adaptive neural network models that translate planned action sequences, an FPGA-based embedded hardware controller, and a smart house interface. Two working demonstrations were delivered — one using a software controller and one using the final hardware controller connected to smart house devices.

Audience

Who needs this

Assistive technology manufacturers developing brain-computer interfacesSmart home companies targeting accessibility and hands-free control marketsFPGA and embedded systems companies seeking medical or IoT applicationsRehabilitation clinics and hospitals exploring next-generation assistive devicesNeurotechnology startups building on brain-machine interface platforms

Business applications

Who can put this to work

Assistive Technology

mid-size

Target: Manufacturers of assistive devices for people with motor impairments

If you are an assistive technology company dealing with the limitation that current brain-machine interfaces only support one-command-at-a-time control — this project developed an FPGA-based embedded controller that reads predictive neural signals and translates planned action sequences into device commands. The system was demonstrated at TRL4 with real neural data controlling smart house devices. With 5 partners across 3 countries and EUR 4,236,000 in EU funding behind it, the core algorithms and hardware design are ready for integration into next-generation assistive products.

Smart Home Automation

enterprise

Target: Smart home platform providers and IoT device manufacturers

If you are a smart home company looking for the next interface beyond voice and touch — this project built a working smart house interface that accepts complex, multi-step commands from a single brain-planning signal. The final demonstration showed hardware-based control of smart house devices using two different test conditions. For smart home companies targeting accessibility markets or hands-free premium segments, this controller architecture could unlock an entirely new input modality.

Embedded Systems and FPGA Development

SME

Target: Companies specializing in real-time embedded controllers for medical or IoT applications

If you are an embedded systems company seeking high-value verticals for your FPGA expertise — this project designed a dedicated FPGA-based controller that processes neural signals in real time and outputs smart device commands. The industrial partner in the consortium specifically assessed commercial potential of the embedded controller and its smart house interface. The controller was validated at TRL4 with 21 deliverables documenting the full design chain from neural data to device actuation.

Frequently asked

Quick answers

What would it cost to license or adopt this technology?

The project was funded with EUR 4,236,000 under an EU Research and Innovation Action (RIA). Licensing terms would need to be negotiated with Georg-August-Universität Göttingen as coordinator and the industrial partner. As a publicly funded project, results may be available under favorable academic licensing conditions, but specific terms are not published in the project data.

Can this work at industrial scale or is it still a lab experiment?

The project explicitly targeted TRL4 — validated in a laboratory environment. The final demo showed hardware control of smart house devices, but this was a controlled demonstration, not a mass-market product. Scaling to commercial deployment would require significant further engineering, clinical trials, and regulatory clearance.

Who owns the intellectual property?

IP from EU-funded RIA projects typically belongs to the consortium partners who generated it. The consortium includes 1 industrial partner and 1 SME alongside 3 universities and 1 research organization. The industrial partner was specifically tasked with assessing translational and commercial interests, suggesting they may hold or co-hold key exploitation rights.

How far is this from a product someone could actually buy?

The system reached TRL4 with a working FPGA-based hardware controller demonstrated with real neural data. However, the objective describes the end-use case — motor-impaired patients planning daily tasks by thought — as a 'far-future vision.' Reaching a commercial medical device would require clinical validation, miniaturization, and regulatory approval, likely placing a product several years away.

Does it integrate with existing smart home systems?

The project built a smart house interface (WP4) that connects the embedded controller to smart home devices. The final demo deliverable describes controlling smart house devices using the hardware controller. Based on available project data, the specific smart home protocols or platforms supported are not detailed, so integration with commercial systems like Google Home or Alexa would need to be verified.

Is there regulatory risk for brain-machine interface products?

Yes. Any commercial brain-machine interface intended for patients would be classified as a medical device and require CE marking under the EU Medical Device Regulation. The project itself operated at TRL4 in a research setting and did not pursue regulatory approval. A commercializing company would need to budget for the full clinical and regulatory pathway.

Is there ongoing support or follow-up research?

The project closed in April 2021. The consortium of 5 partners across Germany, Denmark, and Spain built deep expertise in predictive neural decoding. Based on available project data, no direct successor project is mentioned, but the coordinator at Göttingen and the industrial partner would be the primary contacts for continued collaboration or licensing.

Consortium

Who built it

The Plan4Act consortium is compact — 5 partners across Germany, Denmark, and Spain — with a strong academic core of 3 universities and 1 research organization led by Georg-August-Universität Göttingen. The single industrial partner (which is also the only SME) was specifically tasked with assessing commercial and translational potential, particularly around the embedded FPGA controller and smart house interface. The 20% industry ratio is low for near-market projects but typical for FET Proactive research. For a business looking to license or co-develop, the small consortium means fewer parties at the negotiation table, and the industrial partner's explicit commercialization mandate suggests there is already a pathway assessment in place.

GEORG-AUGUST-UNIVERSITAT GOTTINGEN STIFTUNG OFFENTLICHEN RECHTSCoordinator · DE
DEUTSCHES PRIMATENZENTRUM GMBHparticipant · DE
SYDDANSK UNIVERSITETparticipant · DK
MYSPHERA SLparticipant · ES
UNIVERSIDAD POLITECNICA DE MADRIDparticipant · ES

How to reach the team

Georg-August-Universität Göttingen (Germany) — contact via university technology transfer office or through the project website

Order a report on this consortium See GEORG-AUGUST-UNIVERSITAT GOTTINGEN STIFTUNG OFFENTLICHEN RECHTS profile →

Next steps

Talk to the team behind this work.

Want an introduction to the Plan4Act team or a detailed technology brief? SciTransfer can connect you directly with the researchers and help assess fit for your product roadmap.

Order a report on this topic More in Health & Biomedical