How can you claim in your ft article that the test was ‘wrong’ simply because he wasn’t using the most up to date Tesla system/firmware? Simply put, what about ALL of the vast numbers of existing Tesla’s that aren’t brand new that CAN clearly crash many times more frequently due to the lack of Lidar in the real world scenarios (apart from the cartoon wall) that commonly exist, as shown by Rober.
It's wrong because nobody expects an ADAS system like Autopilot to do this sort of thing (or at least nobody should expect it.) So it's a huge effort to test something that stopped mattering years ago. The video should have said, "You shouldn't expect Autopilot to handle situations like this, though it would be a bonus if it did, so let's look at the current state of the art and compare that."
Comments
Mark
Wed, 2025-03-19 18:23
Permalink
Mark Rober gets it wrong?
How can you claim in your ft article that the test was ‘wrong’ simply because he wasn’t using the most up to date Tesla system/firmware? Simply put, what about ALL of the vast numbers of existing Tesla’s that aren’t brand new that CAN clearly crash many times more frequently due to the lack of Lidar in the real world scenarios (apart from the cartoon wall) that commonly exist, as shown by Rober.
You are not seeing the wood for the trees.
brad
Sun, 2025-04-27 12:21
Permalink
It's wrong
It's wrong because nobody expects an ADAS system like Autopilot to do this sort of thing (or at least nobody should expect it.) So it's a huge effort to test something that stopped mattering years ago. The video should have said, "You shouldn't expect Autopilot to handle situations like this, though it would be a bonus if it did, so let's look at the current state of the art and compare that."
Add new comment