Understand How Autonomous Cars Work by Pretending to be One

Self-driving autonomous cars have been a fairly constant source of debate for decades in theoretical terms, and for years in practical terms. Opponents to the concept make it clear that they don’t trust computers to handle the complicated task of driving, or simply don’t want their driving privileges taken away. Proponents point out the simple fact that, statistically, self-driving cars are safer than those driven by people, and the potential for reducing traffic and pollution is dramatic.

Aside from those rather practical concerns, there are ethical considerations that even veer into philosophical territory. The famous Trolley Problem is often brought up in this case. The Trolley Problem is a hypothetical situation designed to foster discussion about ethics: imagine a trolley going down the tracks, if you do nothing it will run over five people, if you flip a switch for the trolley to change tracks it will only run over one person. Which do you choose? A self-driving car would, presumably, choose to kill the single person (even if that person is you, the owner).

To help people get over their fear of autonomous cars, a team from moovel Lab and MESO Digital Interiors built a small car that you drive while “seeing” the world as a self-driving car does. This works by taking the sensors commonly found in autonomous vehicles (a stereo camera, video camera, and LIDAR), feeding their data into a PC and NVIDIA Jetson TX2, and then displaying it on an Oculus Rift virtual reality headset.

This gives participants the feeling of actually being the self-driving car — at least in a way. As you can see in the video, it mostly freaked the drivers out. Whether this actually helps the cause for self-driving cars or not is yet to be seen, but at the very least it shows people that they’re not operating on some kind of magic.