Soil Analyzer Network


California is the largest economy in the United States, it accounts for more than 13% of its gross state product (GSP) and by itself it's comparable to the many countries of the world. Its agriculture is a very important, multibillion dollar sector that consists of more than 75 000 farms and ranches that have been hurt by the recent droughts. To help fight the droughts modern day farms use large irrigation systems with vast water requirements. Smart management of this precious resource is a must.

Luckily, with today’s technologies we can hack the solution.

What it does

The challenge is to help mitigate California's water crisis by using smart interconnected devices or with what is commonly known as Internet of Things. We have come up with a solution that optimizes water consumption of the irrigation systems with the goal to reduce overall strain on this precious resource.
Our solution consists of two parts, low power soil moisture sensor probes that are scattered across the field, and cloud computing. By using a simple wireless communication protocol, probes aggregate information about certain sectors of the farm's soil and propagate information to Samsung Artik. There, the data is analyzed and used to control the irrigation system, while offering the farmer complete control and overview. Additionally, nodes are based on low power microcontroller that collects the data from capacitive moisture sensor. They are solar powered with battery backup. Data can be propagated from the farthest to the closets node by using a mesh networking algorithm. This setup allows us to build scalable solution which can be custom fitted for a wide range of cases. Our user interface is implemented by using Web 2.0 technologies. We are using Google Maps, Javascript API on the frontend part of the application to show soil sensors as individual markers on the map. They are updated dynamically to show real time information about the field. By using Samsung Artik IOT solutions we can unlock new possibilities and build scalable and complex systems with ease.

How did we built it


  1. Apply for MAD Challenge to get this new powerhouse in small footprint
  2. Buy necessary ingredients such as:
  3. Get five students and one colleague on false pretenses in a room with a sturdy door
  4. Lock them in the room (don't worry, they had access to Internet)
  5. Promise them that you will let them go when they make something fabulous
  6. Evalute their solution and if it's not good GOTO 4
  7. Claim that you did it by yourself and reap the benefits...


On a serious note, we combined a few development platforms and used readily available hardware solutions. We have dissected solar lamps and redesigned power supply (simple boost with low quiescent current), integrated a low power microcontroller into a plastic case with means of RF communication (Artik 1 should fit there perfectly also). The circuit is connected to a capacitive sensor made from PCB traces and stop mask.


For the collector we also used readily available development platforms with network connection capabilities. The boards are running FreeRTOS real-time operating system and lwIP stack. To connect to the Artik 5 we developed a simple information exchange protocol by using HTTP.


On Artik 5 we are running Node.JS with all bells and whistles so that we can collect data and present it in a simple user interface.

Challenges we ran into

Students kept escaping the room through the nearest window. I would like to take this opportunity and say "Thank you" to the makers behind GPS technology...

Accomplishments that we're proud of

Personally I'm proud of my team mates. In a relatively short amount of time we have built a working demo system. Sure it needs fine tuning, more elaborate code and user interface but the really cool thing is that it can measure moisture and show it on a tablet.

What we learned

Time constraints are here as a motivation source.

Also Artik is a capable platform but it needs a community of hackers to become even better.

Solutions can be hacked but only with bright minds of today thinking about tomorrow's problems.

What's next for SAN - Soil Analyzer Network

Networking needs to be developed additionally and the system needs to be tested thoroughly. User interface needs to be prettier and we need to add more functionality to it. Probes power consumption needs to be optimized even more. Capacitive sensors need more accurate CDC (Capacitance to Digital Conversion ) solution and etc. This isn't a product, but a working concept with stable foundations. With additional funds and time this could be the low cost scalable solution.

Built With

+ 1 more
Share this project: