A pair of jeans reminds you that you forgot the low battery Airpods in your back pocket. Or the wallpaper informs you that you did not pick up your credit card before leaving the house. These are all future applications that Microsoft engineers think of for the new technologies being developed in the field of smart textiles.
The goal of this project, called Capacitivo, is to use a recognition technology called capacitive sensing to create fabrics that can recognize the objects they touch. This method is well known among electrical engineers, and it relies on electrodes that detect and measure changes in charge when they are in contact with a given object.
Combining capacitive sensing and AI, Microsoft researchers have been able to create fabrics that can recognize non-metallic objects such as food and liquids and typical household items such as books, bowls or candles. So far, the results are convincing, and Capacitivo can accurately identify 94.5% of the time in a test involving 20 different objects.
see: A guide for IT professionals in robotic process automation (free PDF) (Technical Republic)
Although the system has difficulty distinguishing between different types of liquids-not always successfully distinguishing beer from milk-Capacitivo is still able to determine whether the container is full or empty, and even for small bench plants, whether the soil is dry or wet.
Therefore, using this technology, smart tablecloths can remind users to water plants. Capacitivo can imagine many other applications. “Can trigger the desired action,” The researcher said. “For example, you can suggest a milkshake formula to the user based on the fruits or vegetables that the user detects through the cloth lining in the basket.”
The technology can also measure how much beverage is left in the glass and automatically update your diet tracking application or remind you that you have left an empty bowl of soup on the table for too long.
The Microsoft team has demonstrated the feasibility of Capacitivo in a proof-of-concept prototype, which contains a 12×12 electrode grid, organized in rows and columns, and connected to a piece of fabric. When a non-metallic object is in contact with the electrode, it will produce electrical displacement in the object, and the amount of displacement depends on the material. Electric displacement changes the charge stored in the electrode, thereby changing its capacitance. Therefore, the object can be detected or identified based on the amount of shift observed in the measured capacitance.
In addition, the rows and columns of connecting electrodes can create a larger image by reflecting the overall shape of the contact area of the object on the entire fabric.
The researchers said: “Our method is unique because we combine mutual capacitance and self-capacitance in one package to identify objects based on their material and the shape of the contact area.”
Then, input the obtained data into a machine learning system, which has been trained on various electrode reaction modes of different subjects. As observed in the initial tests, the high accuracy of the algorithm gave researchers confidence that the technology could be used in a variety of useful new applications.
For example, a smart wallpaper can recognize a credit card placed on a table, and automatically fill in the card information when the user first registers or makes a purchase on the website. In the kitchen, Capacitivo can also identify which ingredients are placed on the counter and let users know the order and time of adding these ingredients to the dishes being prepared.
Microsoft promotes this technology as a way to “make the structure aware of the environment in which it is used,” rather than responding to human behavior.According to the engineer, this is what makes this technology compatible with other projects in the smart textile field (such as Google’s Jacquard items, The clothes can be connected to the wearer’s smartphone.
Earlier this month, the search giant Launched the latest jacquard version, This is a smart backpack that can control music and take selfies with a clap strap.
to request modification Contact us at Here or [email protected]