Waymo and Tesla, Same Goals Different Paths to Level 5
Waymo and Tesla both have the same goal toward fully L5 driverless self driving cars, but they differ in approach from a design and engineering philosophy. Elon Musk relies on computer vision systems that use a combination of ultrasonic, radar and camera devices. Waymo on the other hand use a very fundamental device among driverless car makers and that is LiDAR. Elon believes Tesla can rely on it’s sophisticated hardware with AI software while Waymo implements LiDAR at the heart of it’s systems. While Uber has tarnished the image of self driving cars recently, Waymo and Tesla are way more advanced and have a better track record. Unfortunately Uber doesn’t and it also affected their relationship with Volvo for providing them the XC90, a great SUV. The other important part of making self driving cars more successful is the data gathered from road tests.
It now begs the question, with all the data Tesla has gathered what else would it need to attain L5? Does LiDAR make sense for Elon now or are they just going to keep moving in their own way? New research indicates Tesla’s Autopilot ADAS can improve with the use of LiDAR. Elon even said at one time (why he didn’t believe in LiDAR) “Perhaps I am wrong, and I will look like a fool. But I am quite certain that I am not.” Even Tesla partner Nvidia doesn’t believe CUDA alone can be used to achieve full autonomous driving by relying on computer vision.
It sounds logical and that is how Elon Musk usually thinks (First Principles Thinking). Remember when Elon started SpaceX, they actually built a rocket from the ground up. So I feel maybe, just maybe, Tesla is going to do it their way by developing their own LiDAR or similar system rather than purchase it from another company. At this point I don’t know what else Elon has in mind, but if it is going to beat LiDAR, we have to check it out.
Further Reading:
