Re: Robocar--a decade away?

David M. Palmer wrote:
> In article <4786A826.1030004@[EMAIL PROTECTED]
>, Sea Wasp
<seawaspObvious@[EMAIL PROTECTED]
> wrote:
> 
> 
>>Matthias Warkus wrote:
>>
>>>James Gassaway schrieb:
>>>
>>>
>>>>Actually, of the various automated vehicle plans that I've heard of as

>>>>being seriously considered, the one that seems to be the "front 
>>>>runner" requires some sensors be implanted in the road bed.  But its 
>>>>"only" some relatively inexpensive electronics in holes drilled in to 
>>>>the road every hundred (?) or so feet
>>>
>>>
>>>I.e. about 150 device emplacements per kilometre of motorway. That is a
lot.
>>>
>>
>> And very expensive.
>>
>> That ain't gonna fly.
> 
> 
> 
> Construction costs for roads are measured in the millions of dollars per
lane mile.
> <http://www.dot.state.fl.us/estimates/LaneMileCosts/LaneMilecosts.htm>
>
<http://www.arkansa****ghways.com/Roadway/Costs%20per%20Mile.pdf?Record_Number=8>
> (And much much more in cities: $10s of million/lane mile in big cities,
> $188M/lane mile for difficult projects in corrupt+incompetent cities
like Boston.)
> 
> Annual maintenance is of order $10k/lane mile/year.
> 
> A cost of $100/unit * 200 units/lane mile with a 10 year replacement
cycle isn't 
> cheap, but it isn't out of the question either.
> 

	It is, when you remember that most of those "lane miles" aren't 
Federal, they're local or State. For 100 lane-miles of road (a very 
small number for even a county or small city), that's $100*200*100 = 
$2,000,000. Two million dollars. IN ADDITION to all the regular costs 
(that 10k/lane mile per year. Note that these calculations say that 
the cost for the sensors is twice that, so given that maintenance 
still has to happen, you're asking for tripling your maintenance 
budget at least)

	You will get that out of a county for your pie-in-the-sky scheme in 
one way, and one way only: Federal Mandate, with Federal Dollars 
backing it up. And the cost to the Federal government here will be 
public street mileage (a quick google says roughly 3.5 million miles) 
so 100*200*3.5 million, or around seven HUNDRED BILLION dollars. That 
is of course excluding the fact that many of those 3.5 million miles 
are multi-lane roads; if you just need one sensor per 100 feet 
regardless of different lanes, that would be okay)

	And you didn't even INCLUDE the cost of the workers to do the 
drilling, place the sensor, and cover it back up, or the equipment 
they use to do it with.

	And -- as you say -- this is CHEAP compared to building new 
infrastructure.


-- 
                      Sea Wasp
                        /^\
                        ;;;	
      Live Journal: http://seawasp.livejournal.com