On Friday, 20 April 2012 12:45:25 UTC+1, Terry Fields wrote:
> In the scheme I have in mind, the green light system is automatic and
> depends for its information on the gun sensors. However, the system
> needs an 'unsafe release over-ride', and I suggest this is a human who
> watches the pit lane. If there is no car approaching, the 'green light
> man' has nothing to activate, no human reaction time is involved, and
> the pit-stop is as short as can be managed, the signal from the guns
> going to activate the green light.
>
> If, however, another car approaches to make its own pit stop elsewhere
> in the pit lane, the 'green light' man can press a button to over-ride
> the automatic system while that second car is in the 'release safety
> zone' of the pitted car. This human action will involve no extra time
> lost as that is caused by the other car's presence in the appropriate
> safety zone. The 'green light' man, anticipating the safety zone being
> cleared, releases his button as the appropriate time, and the
> automatic system, having already sensed the 'go' signals from the
> guns, activates the green light. Again, this doesn't involve human
> reaction time to an unexpected event, and the pit stop is again as
> short as can be managed although delayed by the second car.
>
> Terry Fields
I would slightly alter your suggestion - have the greenlight man sitting with a "safe to release" button permanently pressed during the pitstop, and then release it only if it is unsafe. That would mean he was more permanently ready to intervene for safety, but still work on the presumption that a release is safe without requiring human intervention. Its a subtle difference, having a button to press in the event of danger vs a button already pressed and ready to release in the event of danger, but I think mildly better.
Matt