This technology enables wearable devices to achieve precise gesture recognition and robotic arm control in complex environments such as intense exercise and underwater environments, opening new doors for fields such as virtual reality, rehabilitation medicine, and industrial rescue.

Direct Pain Point: Motion Interference is the 'Enemy' of Gesture Recognition

Wearable devices such as smartwatches and sports wristbands have long been integrated into daily life, but traditional inertial measurement units (IMUs) have always been unable to avoid a core issue - weak anti-interference ability.

When you run, the natural swing of your forearm produces motion artifacts; When taking transportation, environmental vibrations can interfere with signals; Even simple posture changes can cause device direction deviation due to changes in gravity vectors.

These interference signals either have similar frequencies or larger amplitudes to gesture signals, making it easy to "overwhelm" real instructions. Moreover, there are significant differences in the movement habits of different people, further reducing recognition accuracy and turning human-machine interaction in sports scenes into mere "decoration".

In professional scenarios such as industrial operations and rehabilitation training, this problem is even more fatal. Imagine rescue workers trying to control a drone to explore a dangerous situation while running, but frequently making mistakes due to signal interference; Rehabilitation patients attempt to control assistive devices through gestures, but are affected by body movements that affect instruction transmission - the limitations of traditional devices have always constrained the practicality of human-computer interaction.

Dual innovation: hardware+AI to create an "anti-interference artifact"

In order to overcome this challenge, the research team has proposed a dual solution of "hardware innovation+algorithm optimization", allowing wearable sensors to have both "flexibility" and "smart brain".

Mini stretchable sensor

Thin as paper, comfortable to wear, integrated with a six channel IMU, EMG module, Bluetooth microcontroller unit, and stretchable battery, capable of wirelessly capturing and transmitting gesture signals.

This sensor has a size of only 1.8 × 4.5 square centimeters, a thickness of 2 millimeters, and a stretchability of over 20%. It can be attached to the forearm without affecting movement at all. It adopts a four layer exquisite design, with each layer performing its own duties: the first layer is the battery layer, which can still work stably after 60 charge and discharge cycles, with a Coulombic efficiency close to 100%; The second layer includes antenna matching units and EMG acquisition modules, using a three-layer electrode structure to ensure the stability of signal acquisition; The third layer integrates a six channel IMU and Bluetooth microcontroller unit for signal processing and wireless transmission. In an open environment, Bluetooth has stable signals within a distance of 20 meters, with a maximum temperature of only 27.7 ℃ after continuous operation for 30 minutes. Wearing it for 1 hour maintains a skin temperature of 34.5 ℃, ensuring safety and no burden.