Today I’d like to announce a new project: Full Stack Embedded. Full Stack Embedded is an interdisciplinary effort to lower boundaries for people to utilize, work with and invent new disruptive technologies that they would otherwise have no access to. We’re doing this by creating useful prototypes, documenting them and releasing them on the Internet, where they can be built by anyone. Additionally, we will be offering courses physically in the developing world to empower people in places where they normally would not encounter these technologies to develop inventions of their own. The focus is on embedded processors, which power the Internet of Things, used in full stack solutions that extend from physical sensors to the Internet and everywhere in the middle.
Full Stack Embedded 2016 kicked off yesterday with our team’s first hack sprint. Integration meetings will take place each week, so stay tuned for more details. This week I’ll just be showing a very sketchy outline of our immediate goals in the technical department.
Weather Station Prototypes
This year we’ll develop a prototype weather station that can be used to collect basic meteorological variables outside around the world. That means that it needs to be:
- Low maintenance
Lack of weather observations leaves weather forecasters blind, limiting not only the possibility of issuing warnings that might save lives and property, but also the ability to look back on the climate of a region and how it’s changed over time. Lots of places around the world are missing weather observations, so our goal is to produce a device that can be used to cheaply and effectively reduce these blind spots.
Steps to get there
- Sensors are integrated into a Raspberry Pi (air pressure, temperature, wind speed and direction, humidity, precipitation)
- The device can provide its own power through solar panels
- The device can establish and maintain an internet connection to report its observations
- The device is weatherproof
- Free drivers are available for each sensor
- A daemon can poll the sensors and store their observations in an onboard database
- A web server can serve the observations in a human readable format over the internet
Current nice to haves:
- A REST API makes the observations available as JSON and BUFR
- The device can publish its observations on Weather Underground
- We can push events configured by users to a mobile app
Steps taken and goals for next week
This week we hashed out what hardware we need. In the course of the next week we’ll be ordering the hardware for the first prototype, getting the repository for the web server ready to roll and setting up the data models. We’ll also be looking at potential contacts in Togo, where we would like to have our first workshop next year. A member of our team is from there and that makes it a great place to start! If anyone has suggestions as to who to contact or how to make this better, we’re open to feedback 🙂 More next week!