Completed Computing & AI Brain & Nervous System

PlasticARMPit: Accelerating the Development of Flexible Integrated Smart Systems

In plain English

AI plain-English summary

A flexible electronic nose, worn under the armpit and powered by a plastic neural network, will learn to recognise the chemical signature of body odour. Today’s flexible electronics—smart patches, wearable sensors, bendable displays—are held back by their processors. The microcontrollers available for flexible circuits are too slow and power-hungry for tasks like real-time odour recognition. The PlasticARMPit project replaces the general-purpose microcontroller with a custom-built digital neural network made from plastic transistors. Because the network is hard-wired for a single task and runs calculations in parallel, it can process sensor data quickly while sipping power. If the prototype works, it will demonstrate that plastic neural networks can serve as the standard processing engine for the entire field of printed electronics. That would unlock a generation of flexible smart systems—not just armpit patches, but also food spoilage sensors, wound infection monitors, and environmental gas detectors—that can analyse complex chemical mixtures without a rigid silicon chip. The project is a proof of concept for a fundamental shift in how flexible devices compute, moving from general-purpose to application-specific hardware.

View original technical description
Future flexible electronic devices will be highly integrated, diversified, and interact with the ambient environment by performing intelligent activities such as fingerprint, vein and odour recognition. These devices will require a flexible high-performance, energy-efficient processor, and it is becoming clear that flexible microcontrollers, which are still in research labs, will not meet the computational demands of future smart applications. Hence, there is an immediate need for a flexible high-performance energy-efficient processing engine to deliver these devices. We propose “plastic Neural Networks (NNs)” as the digital processing engine to accelerate the development of flexible integrated smart systems. The NNs are customised for a specific application, capable of operating in extremely parallel fashion to achieve high performance, and consume low power. The project is the first proposal to pioneer digital hardware NNs as de-facto processing engine of the printed electronics world. It will demonstrate this disruptive concept with a prototype consisting of flexible e-nose sensor array coupled with a plastic NN that can be worn under the armpit to recognise the malodour composition.

View the original record at the funder ↗

Related Research

Grants with similar aims, by meaning.

Engineering Fellowship for Growth - Neuromorphic Printed Tactile Skin (NeuPRINTSKIN) (Ext)
MARGE- Mignon Automated Hardware Generation for Embedded Machine Learning Applications
Electronic Nervous Systems on Chip
Magneto-Optically Reconfigurable Photonic Neuromorphic Networks
Micro-Electronics for autonomous neural implants

Original classification

Collaborative R&D

Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.