Intelligent Plant Growth Monitoring System Based on LSTM Network

Wearable plant sensors (WPSs) can effectively monitor plant growth conditions in the presence of microenvironmental parameter fluctuations, which underlines their immense potential in the field of smart agriculture. Currently, the influence of ambient temperature on plant growth is a research focus in intelligent agriculture. However, it is considerably challenging to achieve real-time and precise monitoring of both physical plant growth and the corresponding ambient temperature using simple and efficient methodologies. In this paper, we introduce a dual-mode (tensile and temperature) WPS, comprising a laser-induced graphene (LIG) sensing layer and a polydimethylsiloxane (PDMS) substrate fabricated through laser inducing and gel-transfer processes. Experimental results demonstrate that the WPS exhibits impressive strain sensitivity (1749.8) and a positive temperature coefficient (0.29 × 10 ^-2 °C ^-1 ) within a wide range of strain (0-50%) and temperature (20-100 °C) values. It even maintains stability under low strains (< 0.1%) or small temperature changes (0.5 °C). Furthermore, it has fast response times (87 ms/3.47 s for strain/temperature response) and good stability (4000/25 cycles for strain/temperature). The high-performance WPS served as the foundation for the development of a wireless intelligent plant growth monitoring system, which employs the Long Short-Term Memory (LSTM) network to effectively monitor and decouple the physical plant growth and the corresponding ambient temperature. Our innovative plant monitoring approach introduces a new paradigm in intelligent vegetation surveillance, with promising implications for applications in smart agriculture.