On Wednesday, October 23, 2019 at 3:12:58 PM UTC-7, Andre Jute wrote:
On Wednesday, October 23, 2019 at 1:36:22 PM UTC+1, AMuzi wrote:
https://cyclingindustry.news/shimano...tem-for-bikes/
--
Andrew Muzi
www.yellowjersey.org/
Open every day since 1 April, 1971
I'm familiar with various forms of ABS, from the crude version taken off Boeing passenger jets for the Jensen Interceptor FF* (Formula Ferguson four wheel drive, developed for grand prix cars) which took the better part of a second to respond, to the modern versions. But I've commented before that a really pleasing ABS effect came with a Gazelle Toulouse c2004 (the black crow sleeper -- http://coolmainpress.com/BICYCLINGgazelletoulouse.html -- before the Toulouse went all trendy) which had a disc brake at the front and a limp early model -41 Shimano roller brake at the back which almost automatically stopped the rear overtaking the front on wet downhills, so, while you had to be careful with the on-off nature of the wretched Shimano front disc, you just slammed the rear brake lever to the stops and held it there to stabilise the bike. I actually considered fitting an older, limper roller brake on my Trek Smover in the place of the Shimano IM70 roller -- http://coolmainpress.com/BICYCLINGsmover.html -- which I found too sharp for a rear brake even on a sporting bike, though on the front it was superior to a disc.
Proper ABS on a bike would be superb for cyclists who hold conversations while they ride, rather than fight the controls every moment.
Andre Jute
Experienced
* My own Jensens (I had two, one each time I got fed up with having to keep two Citroen SM to have one ready to drive and one in the the workshop being sorted) were the common 2WD, but once when the air traffic controllers struck in Rome and grounded my plane just when my girlfriend had to be at the Vienna Staatsoper at curtain-up or lose her job to the understudy, I borrowed an FF and drove her and another singing couple over the snowy passes to Vienna, a scary ride, but a fond memory of a private performance as they sang to keep up their spirits.
The real important thing is that an AI self driving car CANNOT actually predict things from any given set of inputs. This means that it depends on its reaction time to prevent accidents - this works well enough for stop and go traffic but it doesn't work for a child running out from behind a car. A man would think, hmmm, there are children present but they are all on the other side of the street and I should watch to make sure that some kid on my side of the street doesn't suddenly appear to run across to his friends. The AI says to itself, "there is nothing in my sensory range."
So a child appearing would be safe ONLY if the reaction time of the computer is rapid enough. And that means that in a residential area it would be limited to 15 mph (22 feet per second) or even less. A van or pickup is about 18 feet long and the response time of a 1/2 second for the computer and the stopping distance of the car would put you at 22 feet on a dry flat road with good traction.
Really this is something that I can analyze rapidly since it is my business.. The case on a freeway is virtually impossible since you can have cars cutting in front of you and slamming on the brakes with less than a carlength in hand. If you haven't already expected it and started braking as they do that you either run into the back of them or have the guy behind you slam into the back of you.