Executive Summary : | Tripura, India's largest natural rubber producer, is facing challenges due to climate change and global warming, resulting in a reduction in latex yield. One of the major threats is leaf disease caused by Colletotrichum gloeosporioides, which releases specific volatile organic compounds (VOCs) during the onset of leaf and stem diseases. These compounds are identified as biomarkers for Colletotrichum gloeosporioides-affected leaf disease. A smartphone-based plant wearable system is proposed for real-time monitoring of leaf diseases using chemi-resistive profiling of menthol and phellandrene. The system consists of three modules: a multiplexed sensor array, a flexible wireless sensing circuit, and a machine learning algorithm-based data analysis and user interface in a smartphone. The key element of the system is a highly sensitive and humidity-resilient multiplexed chemi-resistive sensing array, which consists of two types of hybrid sensors: n-type (resistant with reducing VOCs) and p-type (resistant with increasing VOCs). The hybrid structures of Ruthenium nanoparticles/reduced graphene oxide (rGO) and Ws₂ nanosheets/rGO hybrid structure are proposed for n-type and p-type sensors, respectively. The flexible and stretchable circuit consists of bio-inspired and screen printed wave patterned electrodes and a low-cost wireless sensing circuit, which convert strain into a curvature effect and protect the sensing system from cracks or fractures. The sensing data collected in a smartphone through WiFi will be analyzed by a supervised machine learning model, classifying VOCs even at mixing conditions. This innovative approach aims to provide rapid and on-site early diagnosis of plant diseases and long-term plant health monitoring, especially in resource-poor settings. |