Plant Growth Monitoring, Prediction, and Self-Regulation Utilizing MXene/CNTs/TPU Flexible Strain Sensors Integrated with Deep Learning Algorithms and Soft Actuators
Smart agriculture demands continuous, non-invasive monitoring of plant growth dynamics to enable precise, automated management. Existing rigid sensors and remote sensing platforms suffer from insufficient spatial and temporal resolution, mechanical mismatch with plant tissues, and inability to modulate growth. This study reports flexible, breathable strain sensors fabricated from composite nanofiber membranes (CNMs) of Ti2C2Tx (MXene), carbon nanotubes (CNTs), and thermoplastic polyurethane (TPU) via electrospinning and ultrasonic immersion. The MXene/CNTs dual-network conductive structure yields tensile gauge factors of 5.41, 7.39, and 3.39 over 0–20%, 20–50%, and 50–70% strain ranges, and bending sensitivities of 1.79, 0.89, and 0.46 over 0–30°, 30–90°, and 90–120°, respectively. A tensile sensor coupled with a Long Short-Term Memory (LSTM) deep learning model enables plant growth monitoring and phased prediction. A bending sensor integrated with a shape memory alloy (SMA) soft actuator forms a closed-loop sensing-actuating system that assists leaf growth. The platform demonstrates feasibility for accurate data collection in scientific cultivation and experimental breeding, with potential for unmanned monitoring and regulation in modern agriculture, alpine regions, deserts, or space environments. This work advances smart agriculture by merging flexible strain sensing, deep learning, and soft robotics for plant growth prediction and self-regulation.