Delivery Dynamics

Our hardware/software package is perfect for commercial drone delivery applications.

Comment

Inspiration

We are computer scientists who aspire to work for companies that specialize in integrating smart software solutions to complex hardware ones. Our inspirations include SpaceX, Blue Origins, NASA, and several other aerospace companies. Our passion for this sort of work inspired us to create a solution that we think has several real world implications and is greatly extensible. Drones are an emerging technology and we think that Delivery Dynamics is well worth the effort.

What it does

Our project is a fully loaded computing package for a drone delivery system. We have a web interface capable of placing orders, seeing the drone's flight statistics, and displaying live location updates via google maps. We have built an on-board computing unit that not only sends vital flight statistics to the web interface, but has a verification system using facial recognition as well.

How we built it

For the hardware part of the package, we took advantage of our Intel Edison and the sensors that came with the package to create a great, compact, flight statistics solution. We attached a camera and algorithmically taught it to recognize faces as a user verification step. We programmed the 3-axis accelerometer, temperature sensor, LCD, and a few LED's. We also did a mini-hack where we took apart a keyboard light (piece of swag from GE) and used the LED's to mimic headlights for the drone. The data acquired by the sensors on the Intel Edison sends data over a WiFi connection to Firebase. We then take the data (being updated in real time) and send it to a front-end interface that displays the data. On our front-end, we combine the benefits of bootstrap, jquery, and HTML/CSS to create a simple interface (as proof of concept for scalability) that allows users to see what it would be like to use the drone delivery system, whether it be for medical or commercial use. We use the Google Maps API to show the location of the drone while in flight as well. All of our implementations can be found at our github repository.

Challenges we ran into

  • Integrating facial recognition software with OpenCV, and compiling on the Intel Edison.
  • Implementing a live-tracking function on the Google Maps API.

Accomplishments that we're proud of

  • Our ability to implement several computer science technologies into one package.
  • Machine learning/AI works! We not only use the facial recognition software but program it to learn and recognize the same face. It can differentiate between two or more faces.
  • Implementing a live-tracking function on the Google Maps API.

What we learned

We learned how to work with the Intel Edison and take advantage of its abilities. We learned how to use Firebase and the Google Maps API. We learned how to send the data to a database and pull data from the database.

What's next for Delivery Dynamics

We hope to primarily appeal to Local Motors/Airbus so that they can see the potential in our idea and the sheer potential for scalability. We also hope to appeal to the companies whose software and hardware we used so that we might be able to expand on the idea. We could actually create a professional package with enough funding.

Share this project:

Updates