Irrigation Controller
I developed the Predictive Optimal Water and Energy Irrigation (POWEIr) Controller with a focus on creating water and energy optimal irrigation schedules for farmers in resource-constrained settings. I used model predictive control to optimize daily irrigation schedules based on machine learning predictions of local weather and solar power. Early on in my Ph.D. work, I found that one way to save cost without sacrificing reliability in solar-powered drip irrigation systems is to change the timing of when certain sections of a field are opened to use the available solar power more efficiently. I also noticed that the cost of a solar-powered drip irrigation system was highly dependent on the irrigation volume requirements. I dedicated my Ph.D. work to developing an irrigation controller that predicts and optimizes water- and energy-efficient irrigation schedules.
Stay tuned for updates and publications!
Initial Prototypes
I spent my summer and fall of 2021 building an initial prototype of the POWEIr controller and conducting tests on the roof of the MIT sailing pavilion above the Charles River. I incorporated the learnings from this early prototype into the next version of the controller. In the summer and fall of 2022, I was able to do some testing on a local farm in Concord, MA.
Stay tuned for updates and publications!
User Analysis
Another Ph.D. student in the group spearheaded the user-centered design of the POWEIr controller. I helped with the interviews and storyboards as part of the co-design process. In October 2021 I visited Kenya and interviewed 16 farmers. In March 2022 I visited Jordan and Morocco and interviewed 23 farmers and over 30 stakeholders.
Stay tuned for updates and publications!
