On the first look, self-parking has to do with a computer program which calculates the trajectory of a car and controls the steering wheel. It's realized by algorithms and implemented by AI experts. It's interesting to discuss the details, for example if a GOFAI symbolic approach makes sense, or a modern neural network is the better idea in realizing such robotic car.

The surprising information is, that a self-parking car works a bit different in reality than expected. In a youtube video [1] it was shown that the car is able to find the parking lot automatically. But after the demonstration was over, the real project was started. There was a need for communication between the participants. The owner of the car was clearly proud that the car has done the task successfully. And the other people in the scene which were equipped with cameras had a lot of questions. What we can see in the demonstration was, that the parking maneuver was over after 20 seconds, but then followed a mini lesson about self-parking cars which took 20 minutes.

The assumption is, that automatic parking isn't a technical problem, but can be treated as an educational task. The bottleneck is not how to navigate the car into the lot, but to explain to the audience what the car is doing. Am i right?

[1] Youtube Video, Self-parking demonstration with the 2019 Hyundai Nexo, https://www.youtube.com/watch?v=iH64Ad5yK7s


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.