Artificial Intelligence – the theory and development of computer systems which can simulate and even improve human intelligence. AI is already being used today in many areas of life and is considered a key technology for the future. We want to show you some interesting use cases in our upcoming series starting with Donkey Car.
Donkey is an open source software which allows you to build your own self driving model car. The project was kickstarted in 2017 by DIY Robocars, a community of developers who are working on self-driving car tech and also host races.
Using the donkey software library, a camera, and a Raspberry Pi, almost any R/C car can be turned into a self driving one. But you can also use this guide with a tested hardware and software setup to build a Donkey Car. All the parts can be bought for about $250 on Amazon and it takes about 2 hours to assemble.
How does the Donkey Car work?
You drive the car around a lined track to capture images and steering angles with the camera. The car can be controlled via laptop or smartphone. While collecting training data, the car itself doesn’t do all that much. It basically takes pictures and sends them to an Amazon server and gets servo commands in return. The server is where the magic happens. First it collects the images and driving information from the user manually driving the car around the track. The server records data from a person driving the car, then uses those images and joystick positions to train a Keras/TensorFlow neural network model in software. Once trained, the model can be loaded on to the car and the car should be able to drive like you.
Besides having fun, the goal of this community DIY project is to learn from others, share successes and improve the platform. Furthermore, they have an advantage over big automotive companies: Donkey Car isn’t carrying people so they can try out a lot of things without much risk, which hopefully leads to drive forward fast innovation.