"New cars sold in Europe from 2022 will have to be fitted with systems to limit their speed.
Under new safety rules agreed by the European Union, all new vehicles are required to have “intelligent speed assistance” systems as standard equipment.
The EU rules don’t mandate specific technology for the systems, which can be temporarily overridden by the driver. Some carmakers have already developed ways of using GPS or cameras to detect posted speed limits and make sure vehicles adhere to them."
http://asq.org/qualitynews/qnt/execute/displaySetup?newsID=25434
I'm sure this will eventually come to the U.S(or at least California), so I'm quite interested to see how implementations of such systems and regulations go. My first reaction is that I wouldn't trust such a system to not kill me. It's pretty common to need an extra 10+mph quickly when you're dealing with merging and passing on a busy freeway. Even if such a system can be overridden by the driver, I can imagine it being an issue: if you're not regularly overriding the system, when you suddenly find you need to do so there can be some dangerous delay in doing so. I also worry the system may cause worse problems by drivers becoming accustom to just holding the pedal down till the car stops accelerating.
It is pretty clear that excessive speeding is a major contributor of deaths, and I fully expect a system like this to eventually be a component of solving that problem. I'm therefore quite open to seeing some real-world experiments with such things, but it seems far too early to mandate it. I would love to see it as an option(especially an after-market add-on), primarily to be used on cars driven by teenagers; or even as an enforced add-on for repeated reckless speeders. But a blanket requirement on all cars when we haven't even seen them really tested seems premature. Maybe the article linked above leaves out nuances of the law: for example, if it was a slow-rollout where only some require it the first few years as the system is evaluated, that seems reasonable. But three years is too soon to expect the kinks and dangers to be resolved(or perhaps even discovered). I worry that perhaps the hype of self-driving cars has made lesser, still-complex components seem so clearly "doable" that the regulations are going to push too fast against cost, reliability, and unexpected driver adaptations.
Of course, with speed-related deaths being around 1/4th of vehicle-related fatalities, it's quite possible that even if such a system is flawed it could still result in fewer deaths. This alone could save tens of thousands of lives over the next 10 years if successfully implemented in the U.S. That is a staggering number when the solution could potentially be as straight-forward as comparing GPS coordinates against a database of speed-limits. In such a case, it would be important to move the system to production as soon as possible. I would be 100% behind a near-term requirement for this in the U.S. if it included reasonable conditions for delay or was built using thresholds that help protect the consumers against cost increases. For example, if such a system is sold on 50% of vehicles of the same class in a given average price range(such as within $2000 of final sale-price), it should then become mandatory for all within that bracket the following year. Manufacturer's would ensure they're getting ready to install it, as they wouldn't want to find themselves without a solution and unable to sell cars, while actual deployment is limited by demand and reasonable market conditions. Perhaps better would be to quickly mandate an open interface where an after-market system could be fitted in the future. Then, you could either have it dealer-installed with the new car or added to it later when prices drop. I can't imagine such an interface costing any noticeable amount: just a simple hook into the control system where the outside controller plugs in and dynamically tells the car the max speed to allow, and the car obliges.
