Quote:
Originally Posted by Fuzz
I'm not sure the mechanical bits are in Tesla's. A lot of what I have read is that you really need Lidar to do it properly. Tesla decided not to go that route, and maybe that is why they have issues. Volvo, being from a winter country, has been working with Lidar so they may end up better off the Tesla, despite maybe not having as deep of knowledge as Tesla has with the software side. Lidar is expensive though, and tough to integrate as the units are larger.
|
Well it depends on how you ask. Many people and not just Musk think that the non-Lidar approach is the best.
Quote:
While not as anti-LiDAR as Musk, it appears researchers at Cornell University agree with his LiDAR-less approach. Using two inexpensive cameras on either side of a vehicle’s windshield, Cornell researchers have discovered they can detect objects with nearly LiDAR’s accuracy and at a fraction of the cost.
The researchers found that analyzing the captured images from a bird’s-eye view, rather than the more traditional frontal view, more than tripled their accuracy, making stereo camera a viable and low-cost alternative to LiDAR.
|
https://www.therobotreport.com/resea...-driving-cars/
But, it doesn't mean one is right the other is wrong. The better approach is probably a combination of both, but I wouldn't go so far as to say that the Tesla approach has not worked. From all accounts their system works really well despite Musk being completely wrong on the timelines.