1:8 scale Q2 uses ‘machine learning’ to slot itself into tiny spaces.
The Audi Q2 ‘Deep Learning Concept’ is no ordinary model car. For one, you can’t buy it. And if you could, we suspect the cost would be well beyond the means of your average doting parent. This is because it uses something called ‘machine learning’ to autonomously search for and park itself in scale parking spaces.
Machine learning is exactly what it sounds like – when a machine learns something without being programmed. The Q2 has two mono cameras plus ten ultrasonic sensors – these send signals to an onboard computer, which then sends signals to the steering and motor. This is how it all works, in Audi’s words: “the model car first determines its position relative to the parking space. As soon as it perceives the position, it calculates how it can safely drive to its targeted destination.”
“The system essentially learns through trial and error. To begin, the car selects its direction of travel at random. An algorithm autonomously identifies the successful actions, thus continually refining the parking strategy. So in the end the system is able to solve even difficult problems autonomously.”
All sounds very clever, doesn’t it? Naturally all this know-how is going towards Audi’s real, life-size cars. The model is on display at the Neural Information Processing Systems conference in Barcelona until December 10th.