In nature, organisms have evolved a wide range of sensory abilities to survive. Among these, non-contact sensing—the ability to detect environmental changes and potential threats without direct physical contact—plays a crucial role in the survival strategies of many animals.
Among various non-contact sensing methods, vision—the most relied-upon sense in humans—offers the highest resolution but also consumes the most energy, accounting for over 40% of the brain's perceptual processing power. In contrast to the visual system, spiders have evolved a distinct pathway. Their photoreceptor density is 20 times lower than that of mammals, yet their bodies are covered with an extremely high density of hair-like receptors, reaching up to 400 per square millimeter. These hair-like receptors can convert external non-contact stimuli (such as air currents caused by prey) into neural spike trains, with each event consuming less than 100 pJ of energy—hundreds of times lower in energy density than the visual system. This strategy achieves broad sensory coverage while minimizing energy consumption and overcomes many limitations of visual perception. Such efficient and energy-saving sensing mechanisms are now providing profound insights for modern robotics and artificial intelligence perception system
Inspired by this, a research team from Nanjing University recently proposed a flexible spiking hair sensor (FISH), modeled after spider hair sensors, capable of converting airflow signals into electrical pulses in real time for non-contact sensing. With a power density below 100 nW/cm² and an energy consumption of approximately 660 pJ per sensing event, it is nearly on par with spider hair sensors and reduces energy consumption by two orders of magnitude compared to traditional non-contact sensors.
▍Structural Characteristics of Flexible Pulsed Hair Sensors (FISH)
So, how does this ultra-low-power sensor achieve its performance? The core design of the research team lies in its unique structure. This novel sensor, named the Flexible Hair-like Sensor with Impulse Response (FISH), consists of hair-like sensors based on polyimide (PI) and flexible TS memristors based on Ag/PI/LIG/PI, capable of converting airflow information into pulse sequences for non-contact sensing.
Hair-like Sensor: The Essence of Bionic Design
The hair-like sensors of FISH utilize laser-induced graphene (LIG) technology to create sensing elements with a width of only about 25 micrometers on a polyimide substrate. Scanning electron microscopy reveals that the LIG exhibits a foam-like porous structure, which not only enhances the sensor's response sensitivity but also ensures excellent flexibility.
The sensor can detect airflow velocities as low as 0.4 m/s and, by adjusting the thickness of the PI substrate, can achieve a minimum detection limit of 0.04 m/s. At an airflow velocity of 7.0 m/s, the sensor's response and recovery times are only approximately 40 milliseconds and 26 milliseconds, respectively. After 4,500 cycle tests, the sensor maintains stable performance, demonstrating excellent reliability.
Flexible TS memristor: The Core of Pulse Coding
The flexible TS memristor is key to FISH's implementation of pulse coding. It exhibits typical synaptic behavior: when the applied voltage exceeds the threshold, the device switches from a high-resistance state to a low-resistance state; when the voltage falls below the holding voltage, it spontaneously returns to the high-resistance state. This characteristic enables the memristor to generate self-excited oscillations under current bias, producing voltage spikes with adjustable output frequencies.
The memristor exhibits exceptional stability: after 1,000 repeated cycles, the coefficient of variation for the high-resistance and low-resistance states is only 7.71% and 10.71%, respectively. It maintains stable operation under various bending radii (3-20 mm) and temperature conditions (40-200°C). When the input current increases from 100 pA to 200 nA, the spiking frequency can be elevated from 155 Hz to 2,650 Hz.
Source: Sensor Expert Network