Content Overview

Shortcomings of the existing technology

1. The trade-off between linearity and stability: Traditional underwater sensors (e.g., capacitive types) have a small linear range (only 10kPa) or poor stability (ion-type sensors are prone to water interference), making them unsuitable for high-pressure/turbulent underwater environments.

2. Complex signal processing: Existing neuromorphic sensors require sophisticated algorithms, making it difficult to meet the demands of real-time underwater monitoring.

Highlights of the Article

1. Bionic high performance: Inspired by the fish lateral line system, the design employs micro-magnetic beads (NdFeB-PDMS) and alternating coils to achieve a highly linear response (R²=0.997) to 200kPa pressure, simplifying signal processing with self-excited magnetic action potentials.

2. Underwater Stability: Verified in marine pools/open ocean, resistant to current/pressure interference, with signal fidelity.

3. Precise Application: The game control accuracy rate is 92.19%, and the underwater object recognition rate is 94.71% (Machine Learning + SNN).

Application scenarios

1. Underwater robot: Navigation, obstacle avoidance, and target recognition.

2. Ocean Monitoring: Real-time monitoring of seawater pressure/flow rate.

3. Human-Computer Interaction: Game controls, VR haptic feedback.

summary

In this study, the authors propose a bionic neuromorphic soft pressure sensor designed to achieve stable underwater performance and simplified signal processing. This is accomplished through an innovative integration of micro-magnetic spheres, microfluidic channels, and alternating coil connections. The sensor exhibits self-excitation behavior under applied force, demonstrating high linearity (R² = 0.997) in response to pressure changes, with a measurable range of up to 200 kPa. The proposed mechanism generates distinct magnetic action potentials via its alternating coil design, enabling efficient signal processing. This artificial neural receptor achieves a 92.19% accuracy rate in gaming control applications and 94.71% accuracy in underwater object recognition using machine learning. Additionally, the sensor has been validated in experimental marine pools and open ocean environments, confirming its potential for underwater robotics, marine environmental monitoring, and marine industrial applications.