Ecological sustainability depends critically on the ability of world food production to manage increasingly limited natural resources (e.g. arable land and water) with new techniques that both enhance environmental stewardship and increase farm productivity.  Food security and food safety are critical societal functions, but without significant technological advances, they will be difficult or impossible to ensure in a way that is ecologically sustainable.  To make agriculture more productive and the productivity gains sustainable, growers are increasingly turning to environmental sensor measurement, data acquisition, and data analysis. However to date, these tools have failed to achieve widespread use by smallholder agricultural concerns.  Importantly, individual growers and ranchers are underserved by many recent advances in the commercial and research sectors that make data analytics consumable as simple "black box"end-user products. Current decision support offerings for these constituencies are variously limited, proprietary, complex, costly, require that growers relinquish control over their data,  or are not widely available.

To address the problem of sustainable food security and food safety, the major outcomes of this project are a novel, unifying cyberinfrastructure and agriculture analytics that enable precision, agronomics-driven farming by individual growers unlike any that is available today.  Our system, called SmartFarm integrates disparate environmental sensor technologies into a customized open-source cloud-based data appliance and new analytics that provides growers with a secure, easy to use, low-cost data analysis and decision support system.  Using open-source software and private cloud platforms, this data appliance can be hosted at a range of scales including personal, private clouds on-farm, large-scale public clouds, or in combination (cloud hybrid).

Over the course of this project, we have advanced the state of the art in distributed and programming systems and the Internet-of-Things (IoT) in multiple ways.  We have developed

• a hybrid cloud approach to agriculture analytics for enabling sustainable farming practices. SmartFarm integrates disparate environmental sensor technologies into an on-farm, private cloud software infrastructure that provides farmers with a secure, easy to use, low-cost data analysis system. SmartFarm couples data from external cloud sources (weather predictions, satellite imagery, state and national datasets, etc) with farm-local statistics, provides an interface into which custom analytics apps can be plugged, and ensures that all private data remain under grower control.
• "Where’s The Bear (WTB)", an end-to-end, distributed, IoT system for wildlife monitoring. WTB implements a multi-tier (cloud, edge, sensing) system that integrates recent advances in machine learning based image processing to automatically classify animals in images from remote, motion-triggered camera traps. We use non-local, resource rich, public/private cloud systems to train the machine learning models, and "in-the-field" resource-constrained edge systems to perform classifcation near the IoT sensing devices (cameras).
• a software infrastructure for edge clouds (private clouds located at the network edge), designed to provide reliable, “lights out” unattended operation and application hosting in IoT deployments. Our system implements reliable private cloud operation in restricted resource environments and data durability features that hosted applications can leverage. We leverage it for hosting Hadoop applications at the edge. 
• a scalable, open source, clustering service for K-means clustering of correlated, multidimensional data called Centaurus. Centaurus provides users with automatic deployment via public or private cloud resources, model selection (using Bayesian information criterion), and data visualisation. We apply Centaurus to a real-world, agricultural analytics application and compare its results to the industry standard clustering approach. The application uses soil electrical conductivity (EC) measurements, GPS coordinates, and elevation data from a field to produce a ‘map’ of differing soil zones (so that management can be specialised for each).
• new methods for improving the accuracy of outdoor temperature prediction using small, low-cost, single board computers (SBCs) used in IoT deployments. Predicting temperature without dedicated temperature sensors frees up space on these systems for other sensors and reduces the cost of microclimate sensing (e.g. as used in IoT-based, agricultural applications). Our approach uses multiple linear regression and combines measurements of on-board processor temperature from multiple SBCs with remote weather stations. 
• new distributed services model and architecture for  IoT applications. We hypothesize that IoT devices at the edge are better modeled as servers, which applications in the cloud compose for their functionality. We investigate the implications of this “flipped” IoT client-server model, for server discovery, authentication, and resource use. 
• and, a new distributed runtime system and scheduling service implementing a functions-as-service (FaaS) programming model for IoT. We extend this FaaS model so that it is suitable for use in all tiers of scale for IoT – sensors, edge devices, and cloud – to facilitate robust, portable, and low-latency IoT application development and deployment. We then use the model in a distributed setting to schedule application functions across multiple local and cloud systems. Moreover, the system extends FaaS to leverage hardware acceleration when available to support the next generation of machine learning and statistical decision support applications.

Last Modified: 01/11/2021
Modified by: Chandra Krintz