It seems that not only is Tesla’s more advanced semi-autonomous system better than the automatic driver, but some FSDs are more equal than others
–>
- An owner of Tesla has just repeated the false road wall test of Mark Rober.
- In some tests, they obtained the same result and in others, not.
- The differences seem to be due to more recent versions of FSD.
Earlier this week, Mark Rober triggered a giant online battle by testing autonomous driving technology. In a long video, entitled Can you deceive an autonomous car?He settled Lidar against optical systems like those that Tesla uses. The result? Instant reaction, praise, confusion – with regard to the entire Internet has lost their heads. Now someone else has reintegrated the same test and, without surprise, the results are both familiar and a little different.
In short, Lidar tends to see more clearly and more precisely, in certain situations compared to optical systems. This should not be a big surprise, because after all, it is a high definition radar system that can detect objects in full darkness.
However, when Rober’s video has highlighted Tesla’s failure to detect a wall that looked like a real road, fans of the brand came out with their forks, so to speak. On their credit, the Rober test did not use the entire self-deputy (supervised) but rather the automatic pilot.
Plus: Tesla’s automatic driver breaks through a false road wall while Lidar Lexus stops like a pro
This is where Kyle Paul, an owner of Tesla himself, comes into play. He decided to relaunch the same test with the same general parameters, but this time using FSD rather than on the automatic pilot. He printed his own wall which looked like the real road on which he was seated and led his model several times.
In each test, the Tesla did not see the wall until it is literally a few centimeters from it. As Rober suggested it in an interview, it is plausible that the ultrasonic parking sensors have noticed the wall rather than the autonomous driving technology.
That said, Paul then changed things by calling on a Cybertruck to perform the same test. Interestingly, he succeeded in the test with flying colors by stopping alone each time he was starting to approach the wall. What is the difference? Besides the obvious fact that FSD is more advanced than the automatic driver, the Cybertruck was on Tesla’s latest FSD equipment called HW4. The Y model, a 2022 -year -old model edition, was not as he executed HW3.
Missing pieces
In particular, some in the comments section highlighted the tests that Paul did not do. For example, he did not test the FSD with a model or in the rain – two factors that could offer a more realistic feeling of how the system works in daily conditions.
However, this should at least help calm the noise around Rober’s video. There is clearly a truth in criticism, and those who continue to challenge Tesla’s approach in terms of autonomous driving are not entirely out of the brand.
