The idea brought up in the Martello presentation struck a chord with us, most people don't think about what happens if a IoT type car loses connection, we definitely hadn't. Technology and progress move forward so fast in today's society, and someone has to deal with the dangers that arise from that. We instantly knew we wanted to try our hand at it.
What it does
During the car's tracked trip, there are periods of disconnection from the internet, or what we call "dead-zones". Dead-zone Finder is designed to read in JSON files which detail the coordinates and signal strength of the car. The tool organizes the data through a Java program and outputs a list of all of the internet "dead-zones" into a readable text file. The program uses Elasticsearch and Kibana to help organize and read the data needed.
How we built it
We took the json data set given and ran Kibana/ElectricSearch on a local server on one of our computers as a tool to organize the data in the way we needed it. Then we took the more well-formatted CSV file and used the data in a java program to find the longitude and latitude ranges where internet connection couldn't be reached ie Dead Zones.
Challenges we ran into
We initially had trouble installing and running the Elasticsearch and Kibana software, but this was overcome through the utilization of online documentation. We then had difficulty getting Kibana to interpret our data properly by dividing pieces of information from each JSON node. Our final challenge came with the integration of our results in Kibana with a java program that would parse the data and only keep the relevant pieces of information. We eventually overcame this through persistent debugging.
Accomplishments that we're proud of
We are proud that we were able to overcome the challenges described above. We showed great perseverance through the completion of this project as we ran into multiple hurdles that were overcome through our willingness to experiment and search for different solutions. This is our first hackathon and we are proud that we were able to create a tool we are content with, let alone create one at all.
What we learned
The whole process of building Dead-zone Finder was an incredible learning experience. As a team, we learned a lot about Elasticsearch, Kibana, and its incorporation in real-world applications. We learned a lot about troubleshooting/debugging in Java. Perhaps most importantly, our team really learned about working collaboratively. The most productive time of our hackathon was when we were talking together and getting everyone's input.
What's next for Dead-zone Finder
We would like to see it work on a larger data set where its capabilities to find dead spots could really be tested. Then we would add a visual representation like a heat map of the internet strength in large areas.