A new study conducted at NC State University has successfully combined the art of embroidery with machine learning to create a fabric-based sensor that can control electronic devices through touch. With the rise of wearable electronics and the addition of new functions to clothing, the need for an embroidery-based sensor capable of controlling these functions has become increasingly important. Integrated into the fabric of clothing, this sensor can activate and control electronic devices, such as mobile apps, entirely by touch.
The device consists of two main components: the embroidered pressure sensor and a microchip that processes and distributes the data collected by the sensor. The sensor is triboelectric, meaning it powers itself using the electric charge generated from the friction between its multiple layers. This unique design utilizes yarns made of two triboelectric materials – one with a positive electric charge and the other with a negative charge – which are integrated into conventional textile fabrics using embroidery machines. Rong Yin, the corresponding author of the study, emphasized the importance of creating a three-dimensional structure for the sensor to function properly.
The pressure sensor collects data which is then sent to the microchip. This microchip is responsible for interpreting the raw input and converting it into specific instructions for connected devices. Machine learning algorithms play a crucial role in ensuring the smooth operation of the device. By differentiating between gestures assigned to various functions and filtering out unintentional inputs from the fabric’s movement, the device can accurately control electronic devices. Yin explained that machine learning allows the device to be trained to recognize different scenarios and interactions.
To showcase the capabilities of their invention, the researchers developed a music playing mobile app that could connect to the sensor via Bluetooth. They created six functions for the app – play/pause, next song, last song, volume up, volume down, and mute – each controlled by a different gesture on the sensor. In addition to music control, the device was used for activities such as setting and inputting passwords and controlling video games. The versatility of the device was attributed to its ability to recognize various inputs through machine learning algorithms.
Overall, this study has successfully demonstrated the effectiveness of combining embroidery techniques with machine learning to create a fabric-based sensor capable of controlling electronic devices through touch. This technological advancement has significant implications for the field of wearable electronics, allowing for innovative and interactive functionalities to be integrated into clothing. By utilizing triboelectric materials and sophisticated machine learning algorithms, the sensor can accurately interpret gestures and commands, providing users with a seamless and intuitive interface for controlling electronic devices.
