Using a new manufacturing process, MIT researchers have produced smart textiles that fit closely to the body so they can feel the wearer’s posture and movements.
By incorporating a special type of plastic yarn and using heat to melt it slightly — a process called thermoforming — the researchers were able to significantly improve the precision of pressure sensors woven into multi-layer knitted textiles, which they call 3DKnITS.
They used this process to create a “smart” shoe and mat, then built a hardware and software system to measure and interpret data from the pressure sensors in real time. The machine learning system predicted movements and yoga poses performed by a person standing on the smart textile mat with an accuracy of about 99 percent.
Their fabrication process, which uses digital knitting technology, enables rapid prototyping and can be easily scaled up for large-scale production, said Irmandy Wicaksono, a research assistant in the MIT Media Lab and lead author of a paper presenting 3DKnITS.
The technique could have many applications, especially in healthcare and rehabilitation. For example, it can be used to make smart shoes that follow the gait of someone who is learning to walk again after an injury, or socks that monitor the pressure on a diabetic patient’s foot to prevent ulcers from forming.
“With digital knitting, you have this freedom to design your own patterns and also integrate sensors into the fabric itself so that it becomes seamless and comfortable and you can develop it based on the shape of your body,” says Wicaksono.
He wrote the paper with MIT students Peter G. Hwang, Samir Droubi and Allison N. Serio through the Undergraduate Research Opportunities Program; Franny Xi Wu, a recent graduate of Wellesley College; Wei Yan, assistant professor at Nanyang Technological University; and senior author Joseph A. Paradiso, the Alexander W. Dreyfoos Professor and director of the Responsive Environments group within the Media Lab. The research will be presented at the IEEE Engineering in Medicine and Biology Society Conference.
“Some of the early pioneering work in smart fabrics took place in the Media Lab in the late 1990s. The materials, integrable electronics and manufacturing machines have advanced enormously since then,” says Paradiso. “It’s a great time to see our research return to this area, for example through projects like Irmandy’s — they point to an exciting future where perception and function will diffuse more fluidly into materials, offering huge possibilities.”
To make smart textiles, the researchers use a digital knitting machine that weaves together layers of fabric using rows of standard and functional yarn. The multi-layer knitted textile consists of two layers of conductive yarn knitted around a piezoresistive knit, which changes its resistance when compressed. According to a pattern, the machine sews this functional yarn throughout the textile in horizontal and vertical rows. Where the functional fibers intersect, they create a pressure sensor, explains Wicaksono.
But yarn is soft and pliable, so the layers shift and rub against each other as the wearer moves. This generates noise and causes variability that makes the pressure sensors much less accurate.
Wicaksono came up with a solution to this problem when he worked in a knitting factory in Shenzhen, China, where he spent a month learning to program and maintain digital knitting machines. He saw how workers made sneakers from thermoplastic yarns that would melt when heated above 70 degrees Celsius, making the textile slightly harder so that it can maintain a precise shape.
He decided to try to incorporate fusible fibers and thermoforming into the smart textile manufacturing process.
“The thermoforming really solves the noise problem as it hardens the multi-layer textile into one layer by essentially squeezing and melting the entire fabric together, which improves accuracy. That thermoforming also allows us to create 3D shapes, like a sock or shoe, that fit exactly to the precise size and shape of the user,” he says.
After perfecting the manufacturing process, Wicaksono needed a system to accurately process the pressure sensor data. Because the fabric is knitted like a grid, he created a wireless circuit that scans through rows and columns on the fabric and measures the resistance at each point. He designed this circuit to overcome artifacts caused by “ghosting” ambiguities, which occur when the user applies pressure to two or more separate points at the same time.
Inspired by in-depth image classification techniques, Wicaksono devised a system that displays pressure sensor data as a heat map. Those images are fed into a machine learning model, which is trained to detect the user’s posture, pose or movement based on the heat map image.
After training, the model was able to classify the user’s activity on the smart mat (walking, running, doing push-ups, etc.) with an accuracy of 99.6 percent and recognize seven yoga poses with an accuracy of 98.7 per cent.
They also used a circular knitting machine to create a close-fitting, smart textile shoe with 96 pressure-sensitive points spread over the entire 3D textile. They used the shoe to measure the pressure exerted on different parts of the foot when the wearer kicked a football.
The high accuracy of 3DKnITS could make them useful for prosthetic applications where precision is essential. A smart textile liner could measure the pressure a prosthesis puts on the socket, making it easy for a prosthetist to see how well the device fits, Wicaksono says.
He and his colleagues are also exploring more creative uses. In collaboration with a sound designer and a contemporary dancer, they developed a smart textile carpet that drives musical notes and soundscapes based on the dancer’s steps, to explore the bidirectional relationship between music and choreography. This research was recently presented at the ACM Creativity and Cognition Conference.
“I’ve learned that interdisciplinary collaboration can yield some really unique applications,” he says.
Now that the researchers have demonstrated the success of their fabrication technique, Wicaksono plans to refine the circuit and machine learning model. Currently, the model must be calibrated for each individual before it can classify actions, which is a time-consuming process. Removing that calibration step would make 3DKnITS more user-friendly. The researchers also want to conduct tests on smart shoes outside the lab to see how environmental conditions such as temperature and humidity affect the accuracy of sensors.
“It’s always amazing to see technology advance in ways that are so meaningful. It’s incredible to think that the clothes we wear, a sleeve or a sock, can be made in such a way that its three-dimensional structure can be used for perception,” said Eric Berkson, assistant professor of orthopedic surgery at Harvard Medical School and sports medical orthopedic surgeon at Massachusetts General Hospital, who was not involved in this study. “In the medical world, and in orthopedic sports medicine in particular, this technology offers the possibility to better detect and classify movement and to recognize force distribution patterns in real-life situations (outside the laboratory). This is the kind of thinking that will improve injury prevention and detection techniques and help evaluate and guide rehabilitation.”
This research was supported in part by the MIT Media Lab Consortium.