Scientists cram an entire computer into a single fiber of clothing — and you can even put it through your washing machine
Scientists have incorporated key computing components into a single, flexible fiber that you can run through your washing machine. The researchers hope to one day weave together many of these fibers into a cohesive “fiber computing” network — in other words, items of clothing with smart capabilities.
Smart textiles, also known as smart fabrics or e-textiles, are materials containing electronic components that enhance the features of wearable devices and other products. These could potentially be used to create materials with in-built computing components that can be used in clothing or in woven displays, among other uses.
One of the earliest modern applications of this was the creation of LilyPad in 2007 — a series of sewable electronic components designed to be used in interactive clothing, toys or sculptures.
One of the most significant limitations of smart textiles is that the computing capabilities of individual fibers are severely limited and that they are not embedded with any individual components.
Because these fibers lack components, basic tasks like interpreting biosignals in real time are difficult — and it’s hard to pick up on signals for subsequent data processing.
But in a new study published June 6 in the journal Nano-Micro Letters, scientists fitted sensing, communication, computation and storage into a single fiber strand.
Related: New fabric can heat up more than 50 degrees to keep people warm in ultracold weather
Each elastic fiber also has 60% stretchability and can be put through the washing machine — meaning it could be used to weave practical garments.
The new fiber also enables smart clothing or devices to have better accuracy due to sensing from multiple points and real-time interactions between the computer and the human, the scientists said in the study.
The future of clothing
Each fiber developed in the study incorporated eight devices, including four sensors — a photodetector, a temperature sensor, an accelerometer and a photoplethysmogram (PPG) sensor, which measures changes in light absorption by the skin — as well as a microcontroller, two communication modules and power management devices, the scientists said in the study. Together, these components achieved data acquisition, processing, storage and result transmission.
To monitor how effective the new system was, the scientist integrated four smart fibers into a garment’s sleeve and pant legs and asked one person wearing the garment to perform a series of body weight exercises, including squats, lunges and planks.
Each fiber runs an individually trained neural network — a collection of machine learning algorithms designed to mimic the way the human brain processes information. This enabled the fibers to recognize various actions in real time, including squatting, planking, arm rotation and others.
A single fiber achieved 67% accuracy in recognizing specific movements, while all four fibers working together boosted accuracy to 95%.
“This remarkable improvement underscores the immense potential of multi-fibre collaborative sensing and distributed reasoning, presenting a transformative approach for intelligent wearable systems that seamlessly integrate localized computation with networked decision-making to deliver robust, high-fidelity performance,” the scientists said in the study.
The scientists said these results highlighted the potential performance of a network of fibers working together, but they acknowledged that there are still challenges in optimizing communication speeds, reducing energy consumption and increasing bandwidth.
To scale up a network of these fiber-based computers, the scientists would also need to improve the efficiency of information exchange between the individual nodes, they noted.
Future research, they added, could focus on building higher-throughput, lower-latency communication protocols that are specifically tailored for these types of fiber computers.
