GGrantIndex
← Search

I-Corps: Image-based System for Crop Management

$50,000FY2016TIPNSF

Carnegie Mellon University, Pittsburgh PA

Investigators

Abstract

This Innovation Corps (I-Corps) project is a software application for drones that assists farmers in optimizing crop management. This team's technology provides an image-based crop management Software-as-a-Service solution that leverages drones for aerial image acquisition. By creating mobile friendly flight planning applications, it puts the farmers in control of their fields. As a result, a drone can fly a specified field (up to 250 acres), collect aerial imagery, and the management system will display an updated map with an aerial representation of the current state of the farmers' fields. With the use of a multispectral camera, the proposed system can detect light frequencies from the canopy of the crops. Those light frequencies can be quantified by a NASA derived index (NDVI) where the current health of each acre can be determined. Over time, this I-Corps team expects to see the platform transform into an insight based application that saves the farmer time, money, and possibly even crops. This I-Corps team has created a mobile application that can communicate a flight plan to any off-the-shelf drone. The app allows farmers to define their fields and automatically generates the flight plan on their behalf. This means that a farmer can easily procure the drone through their preferred channel, and get it collecting data on their fields without having to worry about learning to pilot the drone. The app allows fields to be bound, defined, and labeled so an operator can simply click a given field and an optimal flight plan is generated. The flight plan allows the operator to determine at which altitude they wish to operate, and the algorithm alters the flight plan accordingly. The proposed system is in the process of evolving beyond an image mosaicking and flight planning application into an analytics engine that can generate prescription recommendations for the farmers' fields. This application leverages historic weather data, as well as scraping real time weather data, to aggregate heat units in order to predict optimal actions for the farmers, such as harvest. With near real-time imaging, historic weather data, and ground based sensor data the team believes they will be able to more effectively predict and optimize yield. The I-Corps program will allow the team to understand which specific crops they can most effectively service, prioritize their feature development, and elucidate more thoroughly on their go-to-market strategy.

View original record on NSF Award Search →