Re: to dyno not to dyno
- Ya can put out GPS sendors on tracts and on targets to greatly increase
the accuracy to stay on track. i"m a bit studied up on robo-cycles to see
they use straight steering not counter steering and can run lengthwise
on a loose steep slope and not fall down. They use jerk instants of streering
then straighten up to pull this off. i want to race a completely analog
manual antique cycle against a fearless-stupid robo-cycle. Get on ya
bike and go 15 mph and try to straight steer, that is the level of upset forces
that must be countered and controlled with exact throttle and harsh tip
overs then fling ups to catch. Robo-cycle already know how to drift and
riding lenghtwise on a steep loose slope is Exactly same forces as
a race track turn that is not banked or even off sloped. !st came Camel
jockey's and now cars, next in line .... hobot
On 7/15/2013 2:12 PM, john fisher wrote:
If I were implementing it, I wouldn't rely solely on GPS either. Theres been a lot of progress on vision systems since the first Grand DARPA, and I'd look into using something like that too, plus whatever else could be used to determine track position, like dead reckoning with once-a-lap calibration. Its a hard problem, rapidly determining exact position.
But, it would be more fun to allow the riders to only use a manual switch and save all the robotic intelligence for cleaning my house and driving taxis and such. A robot race is only fun for the engineers involved.