Speeding To Become Impossible In UK?

dmearns writes: "Picked this up from comp.risks. It seems the UK government is planning on requiring computer controlled speed limiters on cars within 5 years. The system uses GPS to figure out if the vehicle is in a speed limited zone. The original story is here. I suppose it is inevitable that officials would think of something like this, but I never thought people would accept it!"

Comments

gps reception?

[sith]

What about places where you can't get a good fix on GPS satillites? I know with my eMap, if theres enough tree or building blockage I can't get any satillites. In downtown minneapolis I have no reception at all.. will the cars just stop totally?

Re:gps reception?

[GregWebb]

Being serious, this problem has already hit... Not on speed limiters thank goodness but still interesting.

Y'see, there's some satellite navigation systems that use GPS. Some are rather better than others, some are downright daft. One that springs to mind is a car which was being roadtested and went round a roundabout several times without an instruction to leave. The problem was that the exit it wanted was just after a bridge - so not in satellite reception and so it wasn't ever (as far as it thought) in the right place to get the signal. A stupid system but this sort of problem does exist.

This isn't the only reason why I oppose this though. It's unsafe and will cause accidents. Sound counterintuitive? Think about it.

Firstly, GPS isn't that accurate. Where limits change periodically, some drivers are going to have different limiter settings than others in all probability. Everyone's trying to drive on the limiter so people are going to get irritated by those who aren't - or, worse, not bother to check their relative speed on the assumption that 'the computer will handle it'.

Now, think about a computer which will stop you breaking the speed limit. This removes watching your speed from the routine of driving as 'the computer will handle it' again, so watch for skid marks on tighter-than-anticipated corners. Watch for appalling driving in appaling conditions as speed isn't something people think about any more. And watch for ridiculous overtaking maneuvers from people annoyed that the person in front is going 1 MPH below what their limiter would allow them to. Think that's daft? Watch the lorries and coaches on the motorways in Britain. Almost uniformly limited to 56 MPH in a misguided safety campaign. This, in fact, causes extra congestion, pollution and accidents due to the speed differentials and bottlenecking it causes - most of the traffic wants to go far faster and we now have the daft situation of the majority of the traffic often being in the fast lane. Ahh, note to US readers - British highways only allow passing on the right under normal conditions. Works better, think fluid dynamics.

Anyway, back to my original point there. Because they're all constrained but trying to go as fast as possible, the smallest distinction becomes worth fighting over. Except that the vehicles can't exactly fight hard as they're all limited to pretty much the same speed. So, overtaking maneuvers become long and drawn out. Relatively safe - if irritating - on a motorway, very dangerous on a single carriageway road.

Now, let's imagine the main reason at the moment for overtaking on single carriageway roads - vehicles travelling much slower than normal for the class of road. This might be an abnormal load, a more cautious than normal motorist, a faulty vehicle or possibly a cyclist or horse. If they're to be passed, the passing vehicle must drive into a lane routinely occupied by traffic flowing in the opposite direction. If you're going to safely execute such a maneuver, you should be in such a position for as little time as possible - which might actually mean driving above the posted limit in the name of safety... Yes, these roads carry a tiny percentage of the traffic - but account for an awful lot of the accidents. They need too be considered.

Now, let's imagine you're planning such a maneuver. How would your execution be affected by a vehicle which would suddely stop accelerating? Introduces a rather unwelcome variable into the equation and necessitates a driver staring at the speedo when travelling in what is naturally a fairly hazardous manner. Not clever.

Perhaps we want to discourage such actions altogether? Oops, bad idea. Results in congestion which leads to annoyed drivers. Annoyed drivers aren't concentrating as hard as a rule, so are more likely to make stupid mistakes. They also aren't necessarily thinking rationally so are more likely to go for a gap which they wouldn't otherwise look twice at - again, causing an accident.

Imagine if this system fails. Imagine if you're stuck at 30 MPH in a 70 zone - a dangerous bottleneck and rather likely to be hit, very hard, from behind. Or if you're in a 30 zone and the system suddenly decides to allow 70. You're driving on the limiter safe in the knowledge that it will regulate your speed - except that it will suddenly jump you forwards at quite a rate. Equally unpleasant.

Finally, let's look at the group who cause most traffic accidents - young men. A group which I'm a part of, as it happens. Now, we already have a car culture which regards speed limits and current enforcement methods with contempts, and one which readily modifies any and every aspect of a vehicle in order to enhance its performance. These limiters will get hacked out so fast it's not worth thinking about. Make them compulsary for the annual roadworthiness test? People will do what they already do with exhaust gas catalysts and simply keep two components - one legal, one not - and swap them for that one day per year. Introduce spot roadside tests? They'll become switchable - exactly what happened with lowriders when they became illegal and led to car hopping today.

We might think that it would still be an improvement as at least _some_ would have limiters - but you're assuming a competent job was done of hacking the limiter out, against the years of experience with other computer software. Look at cracked games back in the Amiga days - they often had serious problems due to the work done on them. Now, a badly hacked speed limiter could be very dangerous as there's no guarantee it will be permanently or predictably off, or won't have side-effects. Perhaps emissions control will go out of the window, perhaps reliability will go. Modern engines often run rather close to the limits of the possible and rely on computer control to keep them functional. The respectable guys aren't going to touch this work and I can't see we have much guarantee of competence with the back street hackers.

This is being suggested to feed a rather nasty, media-driven public climate in Britain right now. It is a daft and dangerous measure which is risking a lot to appease a section of the population who haven't thought the ideas through properly and should be knocked firmly and squarely on the head.

As an aside, there's no way of reliably retrofitting this to all pre-legislation cars... expect values of older performance cars to go through the roof if this ever gets in.

Re:gps reception?

[fosh]

>In downtown minneapolis I have no reception at all.. will the cars just stop totally?

Few places in the world are as rural and remote as downtown minneapolis, I don't think anyone in even the UK has to worry about this. :-)

--Alex the Fishman in NYC

Wait till it's cracked

[Carbonate]

I suppose once the system is in place there won't be any more patrollers on the road looking for speeders. So if you crack it you can speed and no one will looking for you. So for once I suppose I'll be on time :-)

Re:Speed

[Pig Hogger]

Reminds me of the old joke of the Vermont farmer in Texas, asking the rancher:

- How big is your land, sonny?

- Lissen, dude, I can jump in my car in the morning, drive it until the sun sets, and I'll still be on my land.

- I used to have a car like that, too...

--

Re:Thank god we overthrew King George III.

[logicTrAp]

Do you realize how inaccurate radar guns are? How is the cop on the side of the road supposed to know that you in your subcompact 4 lanes over was the one doing 11mph over the speed limit and not that 18 wheeler going the other way? If speed kills, why are accident and death rates still dropping after the repeal of the 55mph national speed limit? (When the NHTSA estimated that it would cause 10s of thousands of more deaths) Speed tickets are simply about revenue enhancement - unfortunately, it's not the $75 speeding ticket that nails you, it's the $1000 in insurance premiums you'll pay as you work off your "violation." Frequently the only viewpoint on speeding ever covered by the media is the government line provided by the NHTSA. For the other side of things take a look at the NMA.

Re:They need GPS for this?

[crucini]

I find your airy dismissal quite naive.

They're not going to give out that info to Some Dumb Joe (tm).

Of course not. They're going to feed it to the government in real time. If you think that's unrealistic, look at the Digital Telephony Act and Carnivore. That could enable some interesting opportunities for an alert law-enforcement type (and I'm just scratching the surface):

• A crime is committed at a certain address. Let's interview the ten cell phone users who were nearest at the time. We'll ask, "What were you doing on 15th street last night?" and see if they can come up with an alibi that matches their GPS trail.
• Analysis of the GPS trails will yield some phone users who live in the suburbs but travel into the ghetto periodically. Obviously they're getting drugs. Let's pull them over as they're leaving the ghetto - they'll either be high or carrying drugs. Even a 10% success rate makes it very worthwhile.
• When criminals are caught, their GPS trails can be used as a reference. So if it turns out that serial killers go to the library on Fridays, drive out of town on weekends, and go to Chinese restaurants on Sundays, other phone users who share these traits could be pre-emptively monitored, arrested, or searched. At least they'd be prime suspects if another serial killer starts operating.

Please note that I haven't even touched on abuse of the capabilities by a malicious cop. I've just listed some legal, sensible steps which law enforcement would take with this technology. I also haven't brushed on the commercial possibilites.

This is a dreadful idea

[cje]

"Speed limiters?" Here in the States, we call them "governors" (that is, they govern how fast your vehicle can go, how fast alcohol can be poured from the bottle, etc.) The introduction of these devices into England would inevitably lead to Britons getting into each other's vehicles and greeting each other with inane phrases such as "How's your guv'nah, guv'nah?" "Ah, give us a kiss, guv'nah, and let's burn some petrol up to the guv'nah, and then Jack's a doughnut, we'll be there!"

This would make the British even less intelligible than they already are. :-(

What happens in an emergency?

[RainbowSix]

What happens if my wife or whomever is about to give birth? Then what do I do? Relax and enjoy the scenery, I guess.
--------

Nice Concept--Won't Work

[John Murdoch]

Hi!

My firm is developing a web-based vehicle locator system which uses GPS receivers mounted in a vehicle, coupled with a GIS database. I can tell you that while this sounds technically feasible, it won't work.

A system like this requires two parts: a GPS-based locator; and a GIS (Geographic Information System). The GPS locator is relatively easy--there are commercial GPS chipsets that (depending on technology) can identify your position to +/- 3 meters. The latitute/longitude values can be mapped against a database of road segments--and that's where the problem begins.

With what speed limit data?
Even computer professionals tend to leap to unbelievable conclusions about the ability of computers. This is an excellent case: GIS data is stored in a massive database (the MapQuest.com database ships on 37 CD-ROMs). Look at any digital map--it is made up of bazillions of very short line segments. Each and every one of those segments has to be categorized by the type of road, the street numbers of buildings on the road, the lat/lon where the segment starts, the lat/lon where it ends, and lots more. But there is some surprising information that is not included in that data: weight restrictions, load restrictions (such as haz mats), and speed limits. Adding that information would be a big job--and an expensive job.

Expensive? We're talking about the government here....
Okay--so there has never been a government initiative yet that was dropped for being too expensive. That's not the only problem. Getting 37 CD-ROMs worth of data into storage on your vehicle may be a worthy challenge, but let's assume that the bright lights in government can handle that one too. (Stop snickering.) The big problem here is that geographic information changes--all the time. That new shopping center you drive past on your way to work? You may have noticed the new traffic signal--but did you notice that the shopping center driveway has a street name? Did you notice that the speed limit on the boulevard past the shopping center changed? Did you notice the housing development behind the shopping center, with a dozen new streets?

Anybody working with street-level GIS data gets a quarterly database update. (Which means we're all re-loading 37 CDs every quarter.) Streets get dug up and moved; streets get extended; streets get eliminated; streets get widened; bridges get closed. A quarter's update might include changes to millions of street segments--it is by no means trivial. And typically there is a 3 to 6 month lag between the time a street is changed or added and the time it appears in the database. In order for this system to work, every single car would have to be updated to the current speed limits--and they'd all have to be updated more or less simultaneously. Consider the spring construction schedule on your favorite interstate highway--what happens when construction starts? They'll drop the speed limit to 25 mph, and make all the computers slow all the vehicles down. What happens on the day the project ends--and only 96% of the cars have had their computers updated to show the speed limit on that road segment to be 75 mph again? Answer: 4% of your drivers die. Updating all that data to every single car--simultaneously poses a massive challenge.

And that's not all. With a GPS chipset that supports WAAG (fixed-position, ground-based GPS transmitters) we can determine your position to +/- 3 meters--but the road data frequently isn't that accurate. In the U.S., for instance, all but one commercial GIS data provider uses street data based on the U.S. Commerce Department TIGR data files. Different providers will tweak the data--fixing problems, even comparing the TIGR data to satellite photography (the difference in price between different vendors is generally due to the detail quality of their maps). But even with the best maps you can buy, vehicles sometimes still seem to appear well off the actual road. Not all the time, not most of the time, but often enough that this system won't work. Consider what happens when the onboard computer on a vehicle on I-78 in New Jersey (speed limit 65) decides that it is really on the adjacent service road (speed limit 25, just 15 meters south) and slows down....

As it happens, this level of imprecision drives me nuts. The client is continually having to remind me that GPS and GIS are about displaying information, and letting a human brain do the pattern analysis to determine what it sees. For example: suppose we have a position report that shows a vehicle at 40.638 North/74.8974 West. Is the vehicle on I-78 (speed limit 65 mph), County Route 513 (speed limit 40 mph), or Old Highway 22 (speed limit 25)? The answer is--could be any of the three. (The lat/lon--and it's real data--is an overpass.) You and I can easily answer that question based on our knowledge of other data points, and perhaps knowledge of the specific vehicle we're tracking. (We can see four other data points showing the vehicle westbound on I-78, and we know the vehicle is being driven by a Pennsylvania resident, and we know that the data points were mapped during the westbound rush hour: we can easily deduce that the vehicle is probably on I-78. But all we can do is deduce that--and all we can deduce is that the vehicle is probably on I-78. The driver may well be on the off-ramp, heading for a nearby McDonald's. We have to accept that level of imprecision, because that level of fuzziness is the best we can do.

There is an enthusiasm for technology today that sometimes can let people get carried away. This is one of those times. The developers will get a grant from the government to pursue the idea. And maybe they'll do pilot studies and tests and generate a lot of good PR. And hopefully they won't kill anybody before they realize the limits of this technology, and give it up.