Harringay online

Harringay, Haringey - So Good they Spelt it Twice!

Views: 1352

Reply to This

Replies to This Discussion

I think that article reinforces the point. The accident where the driverless car dragged a pedestrian was not caused by the car itself, but rather another that pushed the pedestrian into the car's path, and probably the cameras didn't even notice it.

In contrast, you read these frequently:

https://www.bbc.co.uk/news/articles/c87rgyyx53ro

In an ideal model, driverless cars would not have accidents. This is not news, we were already modelling these in University back in the 90s (and not even as research, these were typical lab exercises). The cars would not even depend on visual aids, they would likely have both transponders (like all boats have) and in our day and age even likely communicate directly by wireless protocols. The accidents will remain mostly human factor (when a human does something unexpected at such short notice that the car cannot react). And even in these cases, the reaction time of the car would likely be faster than a human driver in the same position.

In actual fact, with a fully self-driven fleet, cars could drive with less distance between them and adapt speed as a collective, to keep traffic flowing. Even making traffic lights redundant.

My car has lidar, radar and cameras. It has already slammed on the brakes twice faster than me (it wasn't even a serious thing, just a car in front breaking harder than anticipated but still plenty of margin).

As in many things, most of the fear is because of lack of knowledge, but once you scratch over the surface, it really isn't that groundbreaking, most of what is being used was already designed and developed 3 or 4 decades ago. What we have now is cheap hardware (cameras, radars and other sensors) and processing power to analyse the input in real time.

Ruben, Rubin AI chips are currently being manufactured and are due for release later this year.

Yesterday, the world's most valuable company unveiled "reasoning" AI technology for self-driving cars:

BBC: Nvidia boss Jensen Huang on Monday announced Alpamayo

.

REACTION time.

Driverless reaction time may be close to a cat's reaction time (c. a 1/25 second).

Also, cats would not drive like humans:

Commuting

Chilling 1

Lapping it up

N.B. simulated cat behaviour that a cat would never do at the wheel. At least is the orange cat is belted up.

It's quite funny all these call for driverless cars and how they are much safer than humans. Be careful what you wish for is all I can say...

https://streamable.com/t7yjtg

Try to get out of the arena shopping centre on a weekend before Christmas at lunchtime. Or anywhere near Finsbury park on an Arsenal night. Drivers can be the same, and with an actual conscience 

MORE on reaction time: humans compared with two animals species:

1 x reptile vs. 1 x mammal

The fastest reaction times of the best human drivers under ideal conditions, may be compared with the reaction time of a snake.

Fast as a snake is, they are at serious disadvantage in a confrontation with an adult cat.

Cats can react in under half the time of a snake. I suggest that the reaction time of Lidar-equipped, driverless cars may be comparable to that of a pussy cat.

In this video, slithering snakes represent the best human drivers, while cats represent driverless computation speed.

N.B. the fearless felines are great at dealing with mice and their night sight and hearing are great, but I make no suggestion they replace driverless computation.

.

OMFG!

RSS

Advertising

© 2026   Created by Hugh.   Powered by

Badges  |  Report an Issue  |  Terms of Service