To that end, researchers from Kyung Hee University and the National University of Singapore, led by Professor Dongwhi Choi and Professor Chengkuo Lee, created a platform that bridges the gap between self-powered triboelectric sensing and AI-assisted human-centric smart electronics.
Traditional wearable sensors often rely on external power sources and suffer from limited signal interpretation capabilities, which restricts their practical deployment in real-world interactive systems.
The review presents a wearable-centered framework covering self-powered sensing principles, material selection, device architectures, and fabrication strategies.
Triboelectric Sensors
At its core, triboelectric sensors operate through the contact electrification between two different material surfaces, generating self-powered electrical signals without external batteries. Moreover, researchers identify artificial intelligence-assisted signal processing, triboelectric artificial synapses, and neuromorphic computing as the critical bridges that transform passive self-powered sensing into adaptive, intelligent HMI systems.
These bioinspired architectures emulate biological synaptic plasticity, including excitatory postsynaptic currents (EPSCs), paired-pulse facilitation (PPF), short-term potentiation (STP), and long-term potentiation (LTP). Furthermore, this enables hardware-level preprocessing of sensory inputs with exceptional energy efficiency.
The reviewed systems demonstrate remarkable capabilities across diverse domains.
In healthcare, triboelectric stethoscopes achieve 97 percent diagnostic accuracy for cardiac conditions through machine learning, while AI-assisted gait analysis enables personalized rehabilitation with patient-specific pattern recognition.
For gesture recognition, wristband-based systems attain 92.6 percent accuracy for full keyboard input, and multimodal smart gloves integrate tactile sensing, thermal feedback, and pneumatic actuation for immersive interaction.
A wearable-centered review framework for triboelectric sensors, covering self-powered sensing principles, material selection, device architectures, and fabrication strategies.
Source: Yoonsang Ra, Sumin Cho, Xinge Guo, Dongwhi Choi, and Chengkuo Lee.
In device control, triboelectric gloves enable real-time robotic manipulation and virtual instrument playing with classification accuracies exceeding 99 percent.
For intelligent transportation, driver-state-aware sensing systems achieve 94.72 percent accuracy in non-driving behavior identification, dynamically adjusting takeover time budgets in autonomous vehicles.
In robotics, multimodal tactile sensors integrated with deep learning enable object recognition accuracies up to 98.1 percent, while soft grippers demonstrate adaptive grasping strategies for complex manipulation tasks.
Real-time Motion Tracking
When paired with AI-driven signal processing, triboelectric wearable sensors unlock transformative metaverse applications, including real-time motion tracking, haptic feedback, and immersive virtual interaction. Additionally, recognition accuracies in those cases exceed 95 percent for athletic performance analysis and 93.54 percent for gait-based user identification in VR environments.
The convergence of multifunctional perception, embodied intelligence, natural interaction, emotional communication, and bioinspired iontronic materials points toward a future where triboelectric wearable systems evolve from simple self-powered sensing units into integrated intelligent interfaces that seamlessly connect humans, machines, and adaptive digital environments.
Click here to view the paper.