Inspiration
Our inspiration is very simple - we are trying to find unsafe drivers by monitoring ones that are speeding. High school and college-aged drivers have the highest representation in speed-related fatal crashes, and the World Health Organization reports that by reducing average traveling speeds by just 5%, can reduce fatal road crashes by 30%. According to the Insurance Information Institute, crashes due to speeding have cost Americans $40.4 billion a year. We are very new to hacking, so using GPS and road data from the Google Maps API seemed like probably the most straightforward way to address this problem.
What it does
Our system is designed to be started manually by a person when they begin their journey by car. After recording the start position, we use Geolocate on Google Maps to record the position of the car at regular time intervals. After a new position has been recorded, we then ask google maps to provide directions between the start point and new position. We use the estimated time from these directions as the "ideal" time it takes to travel between the two locations. If the time interval between the two positions is significantly smaller than the "ideal" travel time, it means that a driver has significantly outperformed google's time estimate and is likely to be speeding. After this calculation step, we set the current position as the new starting point and repeat the process until the journey is over.
How we built it
We used the Google Maps API's Python implementation for the backend of the project. We also use Flask to create a simple demo website with Python.
Challenges we ran into
Our team is made entirely of three Linguists (not CS students) and so creating a website to demo our project was a new experience for us. Furthermore, we first attempted to create the demo using Django but did not have as much success as we would've liked.
Accomplishments that we're proud of
We are proud to have a product that works at any level, since we were basically starting with no prior knowledge.
What we learned
We learned a lot about the Google Maps API provided by Google Cloud and also got some valuable exposure to Flask and Django.
What's next for Speedy Tracker
Our system is very simple in its goal: we are only trying to detect speedy drivers. According to the National Safety Council, excessive speeding reduces the amount of time the driver has to react in a dangerous situation to avoid a crash, so we hope to expand our program to detect rapid acceleration and de-acceleration. These factors seem to be viable indicators of unsafe driving. We also hope to implement a system to relate driving penalties to possible insurance discounts. Our vision is to have the app record all violations over a long period (perhaps a month) and then have drivers with very few / no violations receive a discount for that month or the next month.
Log in or sign up for Devpost to join the conversation.