A Dam-Inspired Liquid-Metal Fractal Microchannel Strain Sensor for Synergistic Hysteresis-Sensitivity Regulation and Multidirectional Sensing.

Flexible liquid-metal microfluidic strain sensors offer high stretchability and a wide sensing range for wearable electronics, yet their performance is often limited by viscoelastic hysteresis and low sensitivity caused by elastomeric substrates and uniformly deformed conductive channels. Here, we present a dam-inspired microfluidic strain sensor integrating a Peano fractal microchannel with embedded micropillar arrays. The fractal channel mimics a meandering riverbed to introduce multidirectional stress components that suppress hysteresis, while the micropillar arrays function as regulating "dams" to dynamically modulate the conductive paths via a strain-dependent gate-closing effect, enhancing sensitivity. The optimized sensor achieves an ultralow hysteresis of 0.51%, a maximum gauge factor of 9.15 at high strain, and a wide strain range exceeding 326%. By reconfiguring the sensor into a rectangular geometry and orthogonally integrating two units, reliable strain direction discrimination is realized. Leveraging this array with a machine learning algorithm, a smart neck posture monitoring system recognizes six types of neck movements with 97.37% accuracy. This nature-inspired design provides a pathway toward high-performance flexible sensors for motion decoding, health monitoring, and human-computer interaction.