A Self‐Powered Neuromorphic Whisker Array With Memristive Encoding for Morphology‐Dependent Wind Tracking

ABSTRACT The ability to discern airflow direction without physical contact is crucial for autonomous navigation in cluttered and dynamic environments. However, the advanced non‐contact sensing systems still struggle to reconcile the power in accordance with the biological systems. Drawing inspiration from the lifetime morphological maturation of rat vibrissal system, we present a sensing‐memristive encoding‐computer processing integrated concept to investigate age‐dependent wind perception. An artificial whisker array is fabricated using a triboelectric nanogenerator (TENG) decorated with artificial fiber, mimicking the morphological variations of adult/aged rat. The self‐powered sensor is directly coupled with a volatile HfO x memristor, which emulates the encoding dynamics of a biological sensory neuron by converting analog wind stimuli into threshold‐based spiking signals. The system is deployed on a robotic car, where bilateral TENG‐memristor units act as a differential pair for wind direction discrimination. The results demonstrate that the adult‐mimetic morphological variations achieve significantly higher directional accuracy compared to the aged‐mimetic in robotic anemotaxis tasks. This effectively replicates the perceptual decline associated with aging in morphological structure. This design not only addresses the high energy consumption of traditional sensory system, but also establishes a physical model linking the morphology to sensing, therefore promoting efficient perception in next‐generation autonomous robotics.