Tesla’s Self-Driving Fails the Wile E. Coyote Test

One of the biggest obstacles, figuratively, facing self-driving cars is the ability to expect the unexpected, quickly identify potential issues, and respond in a well-reasoned manner to produce a safe outcome. One of the biggest obstacles, literally, facing self-driving cars are giant Wile E. Coyote-style walls painted to look like the road ahead in an attempt to trick them into crashing. Okay so the latter is pretty unlikely to arise in the real world, but that didn’t stop former NASA engineer and current YouTuber Mark Rober from seeing just how well self-driving vehicles stand up to the Looney Tunes test. In his most recent video titled, “Can You Fool A Self Driving Car?” Rober pits two different autonomous vehicle systems—Tesla’s computer vision-only Autopilot and an unnamed system that uses Light Detection and Ranging sensors—up against each other in a series of tests that culminates in an attempt to stop a car in its tracks using the same technique that Wile E. Coyote tried to us ... Read full article.