In my last article, I wrote about a hobby of mine: translating physical objects into digital experiences as a way to test my skills. Like most interaction designers these days, I am fascinated by understanding how digital can connect to our physical reality.
Working with hardware enables designers to create more meaningful, situation-specific interactions that might feel cumbersome or be impossible with software alone. Last year, I had the opportunity to work on a project with IDEO.org (a nonprofit design organization that launched out of IDEO with the mission to improve the lives of people in low-income communities through design) that tested my software and physical design skills. The brief: Apply low-cost sensor technology to precision agriculture for smallholder farmers in developing countries.
Before heading to the field, our small team started by researching precision agriculture practices and technologies for farms and households. We found inspiration in devices like Parrot Flower Power. Even so, we really had no idea what a good precision agriculture device could be, or what challenges farmers faced. But at IDEO, we believe that you learn best by doing. So we quickly ginned up three devices to facilitate conversation with farmers: the Magic Stick, the Ambient Sensor, and the Rice Shoot.
Physically, the Magic Stick was inspired by traditional farming tools like rakes and shovels. The tip of the stick was a temperature/humidity sensor that could be stuck into soil or crops that have been harvested and put in storage. Digitally, the Magic Stick created a data table that combined temperature and humidity with GPS location.
The goal of the Ambient Sensor was to help prevent the growth of a common toxin that spoils crops in storage in many parts of the world. The sensor tested the environment for the conditions that the toxin thrives in: high humidity and temperature. We designed the device to hang on a storage shed wall, and if high humidity and temperature conditions were detected, LED feedback would alert farmers and prompt them to ventilate the space.
We designed the Rice Shoot to measure the water content of rice fields, making the process of rice paddy irrigation more resource efficient, and potentially increasing yields and reducing disease.
Meeting with farmers in Myanmar's Irrawaddy Delta.
We brought our early prototypes to many farms in Kenya, Tanzania, and Myanmar—countries where IDEO.org had existing relationships. During our travels, we spent our days designing, wiring, and programming in the back of bouncing trucks, tidal river boats, and turbulence-prone airplanes. All the people we spoke with were smallholder farmers who need to farm in order to live. Together, we discussed the challenges of growing specific crops in specific regions, as well as any personal challenges they faced. In northern Myanmar's dry zone, we learned that there were no consistent optimal irrigation practices for betel nut, a cash crop that is a big part of the region's economy. Each farmer was watering at different times of day and for different lengths of time, with different yields.
From left to right: the Magic Stick; optimal moisture sensor; the Betel Meter outside its case.
Our time researching with these farmers showed that what they really needed was an upgraded Magic Stick—something with a brain. So we created the Betel Meter, a device that helps farmers create optimal irrigation for betel nut.
The Betel Meter is planted in key positions in a field and takes moisture readings once an hour. Once a day, a farmer presses the meter’s button. A green LED signifies that soil has a healthy moisture level and the farmer can irrigate normally. Red LEDs signify that the soil has been too dry, and farmers should keep irrigation taps open for an additional five or ten minutes. Blue LEDs signify when soil conditions are too wet.
The Betel Meter's "brain."
To build the Betel Meter's brain, we used a TinyDuino from TinyCircuits, an Arduino compatible micro-controller with the power of an UNO. The TinyDuino, which is about the size of a quarter, has a brilliant architecture that allows any additional functionalities beyond basic processing to stack on top of each other. This allowed us to keep the core device small, even though it has a real-time clock, a data SD logger, and GPS boards.
Prototyping the meter's logic.
Creating the baseline data for the meter turned out to be one of the biggest challenges of the project. Because optimal irrigation for betel nut in an agricultural zone as specific as northern Myanmar is unknown, we had to rely on white-paper research and some intuition. We first researched ideal moisture conditions for all plant life. Then we normalized evaporation rates for water in sandy soil, and cross referenced it against average temperatures for each month in the region.
In the field (literally).
The physical design was equally tricky, since the meter needed to withstand extreme weather for long periods of time to protect the sensitive electronics inside. Luckily, we had IDEO.org's Design Director and Co-Lead of the San Francisco studio, Adam Reineck, on the team. He quickly designed a 3D model in Solidworks and printed the form in nylon, which stood up against heavy rain—and even a sandstorm.
Our early prototypes used USB power banks; however, we knew that the Betel Meter had to be an “always-on” device with a reliable and constant power source. We opted for solar power, something the team had experience with. We began by hacking panels out of old solar camping lights and adding a few resistors and extra batteries to the circuit. That way, we could ensure that the device would store enough energy to power the micro-controller and components through the night.
Data output shows irrigation spikes.
Once we had our prototype, we returned to Myanmar to field test the Betel Meter, planting four of the devices in two fields. Not only did the Betel Meter provide irrigation feedback, it also recorded and created what we believe to be the first-ever dataset around optimal soil moisture for betel nut. People from all over the region came to see and take photos of the meter, and we were continuously asked to come back with the final working product.
From here, our ideal next step is to add machine learning that allows the device's software to evolve from its own data.
Working on the Betel Meter changed my personal approach to design and how I attack problems. Now, on every project, I try to find opportunities to create products that reach deeper into people’s physical lives than pixels on a screen.
Working with such an amazing team, in such a great place and on a product that challenged me the way the Betel Meter did, changed my life. I will be joining the team full time at IDEO.org in New York City this fall, where, hopefully, I will be part of the next chapter of the Betel Meter.
Read more on Fast Company