Whether they admit it or not, many people find it hard to park their vehicle. Be it perpendicular or parallel parking, there are drivers that can’t seem to slot their car in perfectly. Not to fret as Audi might just have an answer.
The ongoing Neural Information Processing Systems (NIPS) conference in Barcelona, Spain, is the venue where Audi presented its latest research vehicle, the Audi Q2 Deep Learning Concept. The 1:8 scale model aims to show how technologies like Artificial Intelligence can allow automobiles to perform tasks like parking.
Using a pair of front and rear-facing cameras, along with a bevy of sensors positioned around the vehicle, the Q2 Deep Learning concept vehicle can easily detect where it is and where it wants to go. It begins by determining its current location. This is then followed by a “trial-and-error” process to find the most efficient route it should take to get to its target. In this case, it moves around a 3x3 meter area with a metal parking surround. Once it figures out how to get to the spot (with lightning-quick precision, mind you),, it will then proceed to drive itself.
The Audi Q2 Deep Learning Concept is just the first step in the quest towards making vehicles as intelligent as possible. The goal is to assist humans in tasks like driving by providing a safe, stress-free “autopilot” feature that can take over at any time. With this new technology, along with Audi’s Drive Pilot program, it looks like self-driving Audis are just a few years away.