Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Tesla Sensor Suite vs. LIDAR

This site may earn commission on affiliate links.
If that is the question, than my answer is "We don't know the answer to that". But I am confident that Tesla has figured that out. Public road spaces often contain radar signals, from police radar units and increasingly from other cars and not just Teslas.

Elon said that each radar pulse has a signature and that is how they track the returns. I assume that each Tesla creates a different signature so that it knows to only pay attention to its own pulses.
 
From LIDAR industry, so expect bias:

Lidar: Pressure mounts on Tesla to adopt new Autopilot technology

"Lidow also criticises Tesla and Musk for not using lidar, which he says would have prevented a death which occurred earlier in 2016 when a Tesla using Autopilot failed to avoid a collision.

"Lidar gives you direct, instantaneous 3D images. The processing requirements, latency and margin for error are all much, much lower than with radar," Lidow said. "A laser beam is fired all around you and the time taken for the light to return is measured. It is accurate to a resolution of a couple of centimetres, each pulse of light takes just a couple of nanoseconds and there are a billion pulses per second."

The typical lidar system for a self-driving car, which sits on the roof and spins around to capture an image of the vehicle's environment, has 64 lasers each emitting a pulse of light one billions times every second. "There is not much computation needed to work things out" Lidow said, regarding the images created by lidar. He also said claims lidar cannot see through rain and fog are untrue and the technology's vision through bad weather is "better than humans."

Where radar is good at detecting objects and relaying how far away they are, it cannot determine shapes accurately; lidar can, making it suitable for spotting cars, trucks, bikes, pedestrians and wildlife at the roadside and in a vehicle's path.

Speaking of the crash in May, which resulted in the death of Tesla Model S owner Joshua Brown,Lidow says that lidar would, without question, have prevented the collision."
 
I would like to know what they will say when a crash occour and a Lidow is mounted, since it WILL.
There is no "if"
"Oh, yes, our system saw the big truck, but the software didn't realize the danger of it, so we happily crashed against it"
 
Another Tesla with LIDAR sighting in Palo Alto near Tesla’s headquarters today, Friday, 09/23/2016.

The license plate seems to have the vertical abbreviation "MFG" on the left edge for manufacturers.


Tesla Model S prototype with Velodyne Lidar sensor spotted in Palo Alto

tesla-model-s-prototype-with-lidar-sensor.png
 
  • Informative
Reactions: lklundin
Just like google cars, "the more, the better" and to the pit of doom the costs :D
But then, who cares? they can try using all that stuff and the, with the recording ask himself: ok, now, given our data, what can be safely eliminated? if you discover that the radars got all the [usefull] signals that the lidar get, than you know that that lidar can be eliminated safely.
After all the numbers of leaf on the tree doesn't help much, so even if a lidar catch it, it's a useless info for the navigation system, and if the radar pick up the same trucks / cars / etc that the lidar get, you discover that after all the lidar is not so usefull.
But then, on the other hand you can demostrate that a lidar is essential, and not by word, but by fact. ( of course you should use the way the radar is used in tesla, not the basic mode, else it's useless the comparison ), test that we suspect telsa is doing with the mule equipped with lidar