I-Corps: Developing a Robust and Automated Insect Detection System for Connecting Farm Communities

Award ID: 2331554

The broader impact/commercial potential of this I-Corps project is the development of software technology to provide farmers with data-driven recommendations based on real-time sensory data analysis from their farm fields. The proposed technology also will introduce the concept of Smart Connected Farms (SCF) to enable privacy-preserving data sharing among farmers and service providers via wireless technology. The goal is to disseminate real-time information across different farms and make precise decisions like analyzing insect population growth. Implementing an SCF network in rural farms has numerous networking and technological challenges, including the size and heterogeneity of transmitted data, low/no bandwidth communications, and data privacy. The proposed technology employs low-cost communication medium and edge devices to provide better cost-benefit tradeoffs to the farmers. In addition, data privacy is addressed with a privacy-preserving distributed approach that offers secure and personalized data collection and sharing to eliminate revealing information about the crop quality with a GPS location. The proposed technology also involves the domain experts like agronomists and entomologists to provide precise recommendations. This I-Corps project is based on the development of a platform for rural connectivity and community decision-making integrated with privacy-preserving distributed data analytics. The data collected is used to train machine learning models. Compared to existing technologies, the proposed platform offers a location-based fine-grained prescription for the precise application of insecticides for a novice farmer, reducing farming cost. The prescription is designed to provide insights about a precise infected area, the severity of the infestation, and the number of insecticides to be sprayed. In addition, it may be used to establish a farm network using Long Range (LoRa) communication technology across the field, making it suitable for areas with no network to transmit on-field data to edge devices. The proposed technology addresses the user's issues of adaptability, trust, data security, privacy, and risk preferences to justify the benefits to farmers in collecting and processing real-time farm data. The proposed technology may be applied to developing a broad range of cyber-physical systems and smart services. Such systems may benefit crop consultants, agronomists, data scientists, and entomologists to collect on-farm data and accomplish various applications such as insect detection to recommend insecticide spraying decisions in real time. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Program Director: Ruth Shuman