Inspiration
With the growing popularity and relevance of electric vehicles (EVs) in society, greenhouse gas emissions into the atmosphere are being reduced greatly. Even with this, it is still important to continue to raise the bar, and optimize wherever possible. One commonly looked over aspect of EVs is their unseen carbon footprints. While they have zero emissions in use, they still need to be fueled by energy at charging stations. Charging stations source their energy from various power plants, including harmful sources such as coal and fossil fuel plants. These dirty energy sources are sneaking their way into even the most environmentally conscious people's routines without them even knowing that. Our goal is develop an advanced navigation system layer on top of google maps that will optimize routes to prioritize charging stations who source their energy from clean sources using custom algorithms and machine learning.
What it does
We developed a web application with a UI which accepts one of many supported electric vehicles as input (each has their own custom battery life and efficiency), as well as a start and end location, and displays the optimal route of transportation for that vehicle. Our system displays the estimated time of travel, as well as the calculated carbon emissions for the trip. We also wanted to allow users to compare our routes and emissions with alternative means of transportation including gas cars and public transit.
How we built it
We built this web app on top of a Flask server which we created in python. The flask server contains API endpoints which can be called by our front end to calculate the different values and routes we want to display. Our front end is written from scratch in HTML, CSS, and JavaScript. We calculate the carbon footprint that an EV would inherit from a charging station based on the nearby powerplants and how they produce energy. From this we can create algorithms to select the combination of charging stations which minimize CO2 emissions and time.
Challenges we ran into
Our biggest challenge we ran into is the time limit. Even after working through almost every hour of the competition, we still struggled to finish this project. We spent alot of time trying to implement an ML model to create routes, but were unable to get it working in time. On top of this, our navigation algorithms can definitely be improved upon with more time. We also ended up running over our 300$ free trial limit with googles mapping related API's. We found it hard to try and limit how many calls to the API we could make during our testing, which ended up being a limitation on our progress.
Accomplishments that we're proud of
We're very proud of the depth we managed to achieve in this project with the time limit and a group of only 2 members. We considered many different aspects into this optimization problem that relies on so many different factors to perfect. We were able to implement a dataset of a very large amount of EV's and their own custom data values, as well as were able to track and display carbon emissions of various travel methods. We feel our MVP does our problem statement justice, and while there is room to improve, we reached everything we originally planned to achieve throughout the weekend.
What we learned
This was our first time working with Flask, which was a lot of fun to learn. More importantly, we learned how to organize an with many different components and opportunities into a concise project for the deadline. While we had many ideas flowing and wanted to go in many different directions with the problem statement, we were able to set a list of goals for the weekend and followed it closely. With this we managed to reach our goals, and had the chance to explore more in depth solutions including attempting to create a reinforcement model capable of making predictions.
What's next for EcoNav
Firstly we would like to improve our ML model to explore potentially quicker and cheaper ways of calculating routes. Past this - we would like to improve our optimization by taking climate and elevation into account as well. These are two factors that play a big role in the efficiency of EV's but would add a very high amount of complexity to the project and will require much more time to implement. From there we would like to explore potentially reaching our target audience Environmentally Conscious Commuters (ECCs). We will do this by developing a mobile application or potential feature to be adopted by a large social media platform. From the research we conducted, we learned our target customer is from younger generations and cares deeply for the environment. We would like to continue to tailor to this customer by adding a gauge that can change how much a customer prefers a fast route or an energy efficient route, opening up navigations to longer but cleaner paths.
Log in or sign up for Devpost to join the conversation.