‘Smart tablecloth’ can distinguish between types of food and drink
Image credit: Dreamstime
Researchers from Dartmouth College have led the development of a smart textile (“Capacitivo”) which can accurately detect non-metallic objects.
Although there are already smart fabrics for sensing objects, these typically rely on inputs such as user touch. Capacitivo uses a “implicit input” technique; this means that it does not require any movement from the object it is sensing.
The design prototype features a grid of diamond-shaped electrodes made from conductive fabric and attached to a cotton sheet.
Capacitivo identifies objects through subtle shifts in capacitance in these electrodes. When a non-metallic object is in contact with the electrodes, the field applied from the electrodes causes an electric displacement within the object; the amount of displacement varies depending on the permittivity (an electrical property) of the object. This displacement shifts the charge stored in the electrodes, resulting in a change in capacitance.
The system can detect and recognise an object based on how much of a shift in capacitance is measured, having been trained to recognise patterns associated with various objects.
“This research has the potential to change the way people interact with computing through everyday soft objects made of fabric,” said Professor Xing-Dong Yang, senior researcher for the study [PDF].
20 everyday objects of various materials, sizes, and shapes were tested on the “smart tablecloth”, including cheese, fruit, lipstick, and a potted plant. The system achieved an object detection accuracy of 94.5 per cent. It was particularly accurate at distinguishing between types of fruit, but performed less well for objects which do not create a firm “footprint” on the fabric, such as a credit card.
In a second study, the system was shown to be able to distinguish between different types of liquid - including water, milk, cider, and fizzy drinks - with high accuracy.
“Being able to sense non-metallic objects is a breakthrough for smart fabrics because it allows users to interact with a wide variety of everyday items in entirely new ways,” said Dartmouth PhD candidate and lead author Te-Yen Wu.
When an object or the status of an object is identified by the fabric, it can trigger a desired action; for instance, it could send reminders to water a potted plant when necessary. The researchers hope that the system could have a range of applications, such as finding lost objects, providing information to other smart systems (e.g.: diet trackers), and assisting home cooking by providing recipe suggestions and preparation instructions.
Sign up to the E&T News e-mail to get great stories like this delivered to your inbox every day.