Audi plans to continue expanding autonomous car technology by introducing more features to its automated systems. In order to test out these complex concepts, it helps to initially work on a much smaller scale.
According to Audi, “1:8 scale model car – the ‘Audi Q2 deep learning concept’ – is demonstrating an intelligent parking process. On an area measuring 3 x 3 meters, it autonomously searches for and finds a suitable parking space in the form of a metal frame, and then parks itself there.”
The sensor technology demonstrated in the toy car’s design uses two mono cameras and ten ultrasonic sensors. The built-in computer processes all the received information and converts it into control signals. The converted data is then transferred to the steering mechanisms and the electric motor. This results in the model car steering independently, driving forward and in reverse, and maneuvering successfully to the targeted destination.
Deep reinforcement learning supports the model car’s parking ability. This means the algorithms autonomously learn through each success and failure.
Audi explains this trial and error process: “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.”
This feature is still in the pre-development phase, and Audi hopes to soon transfer this intelligent parking system to a full-size model. You can see the tiny car strut its stuff in the video below!