In which I rant a little about the glaring lack of common sense in
today's navigation applications, end up rather late and stressed for a wedding I should have been half an hour early for... and I explain what's happened to the
sat-nav group test that we promised six months ago...
Read on in the full article.
Sounds like you have been having as much luck as I have.
Ive found that the majority seem to think that simply releasing a version for symbian is enough.
I have some educated guesses as to which company applies to which situation you have given.
However, there is one company that does seem to be taking an interest.
Here's a Satnav following dude I came across in the fields behind out house, that bit doesn't look to bad , but to get there he had to drive across 300m of 4inch deep ruts, but Tomtom does have it as a road!
I've been using TomTom on my Symbian phones since I had my 6600 (So I'm guessing that was version 4 or 5) and I'll be totally honest with you, it has never let me down. My main reason for purchasing it (outside my obvious and rampant gear/gadget lust) was that I had just started a job that required me to travel a lot by car, meeting clients for multiple appointments every day. Now, at this point, I will say that I did have some knowledge of the road network as most of my travel was within East Anglia, where I have lived and worked all my life. However, many of the places I ended up going were new destinations to me. Only once do I recall being sent on what I thought may have been one of those legendary "Sat Nav" trips where you end up wedged between two buildings or lost in the middle of a quarry, but it actually turned out to be a very small and infrequently used C road that actually saved me time ! Suffice to say, the TomTom software was an incredible investment and I don't recall ever being late or lost.
My current job has me travelling the length and breadth of the UK, venturing into unknown territory, and whilst I have fewer "time critical" appointments, TomTom predicts an arrival time upon my departure and I rarely arrive any later. My only beef has been with up to date mapping, but knowing the length of time it takes to produce these maps, and the ever changing road network, this can be allowed for, to a degree.
As you say, one should never rely on Sat Nav alone. To do so would be utterly foolish. The amount of variables involved in a journey are so many that no currently available Sat Nav could allow for them.
And therein lies your issue. What you are asking for is artificial intelligence within an environment that is, for the most part, controlled by human nature. It's chaos theory, to be honest. The old adage, "The flap of a butterfly�s wings in Brazil set off a tornado in Texas", is very pertinent here.
An individual gets up in Chobham one morning and unbenkownst to him, his momentary laspe of concentration at the wheel that causes a minor shunt on the A322 is what spoils your day on the way to Newcastle on a Saturday morning. No piece of Symbian software will currently allow for accurate forecasting of that nature, sadly. Not even super computers can predict human nature and it's free will. Sure, you can probably plot a characteristic on any given road, stating that at certain times of the day, average speed is affected by "so much", but it's never the same amount and the possibilities are endless. think of the amount of data collection and calculation needed to come up with your UCA rating. Don't get me wrong, if it could be done, I'm with you all the way, but the effort required would probably, no, definitely incur vast costs which would only spill down the the end user.
No, I'm afraid that good planning for travel, at this stage of technological development, involves more than just relying on a Sat Nav device and requires a combination of tools from physical map reading prior to the journey to tuning into local radio for traffic announcements and above all, sheer bloody luck.
This should not, however, deter us from craving such things and pushing the envelope so that our kids may benefit from it (or their grand kids, given the current state of progress with TomTom on Symbian !)
An interesting debate 😊
It moves the topic away from Symbian somewhat (at least for now, but there is no practical reason it can't feature on a Symbian Device in future).
But what you are ACTUALLY talking about Steve, is not so much Sat-Nav "Artificial Intelligence", but real driver "Actual Experience History".
Enter, TomTom IQ Routes...
Explanation: The now being released TomTom x30 Series of standalone PND devices, include a feature called "IQ Routes". In reality what this actually is, is HISTORIC, real world, user supplied data from us lot that have been giving this to TomTom anonymously without realising what they were building it all up for, these past however many years.
It is actual data for time taken to travel routes and roads, and TomTom uses this instead of a simple stated speed limit value for the roads in question.
Thus, depending on the day of the week, weekend etc, when you travel, different routes may be supplied - in your example Steve, on a quieter weekend, with perhaps less traffic on the main motorway, IQ Routes perhaps, for SOME motorways, knows that these are still the quickest route.
However, rush hour during the week, the real actual historic data it has will tell it that the motorway is log-jammed and will take forever, and thus local roads and junction hopping etc will be quicker.
That is the THEORY anyway... So far it's way too early to report on how well or otherwise this works. But at least the PRINCIPLE is sound.
Certain short-comings for now even with this approach, is that it only lets you specify the DAY to be travelled, not the time, so can't differentiate between a rush hour jam, and a mid morning flyer.
But it's a START towards reaching the level of proper reliance on sat-nav that your article talks about.
Couple this with TomTom HD Traffic - again real-time traffic info supplied by users that actually have a special vodafone sim card and data connection with the device, and you can measure the REAL time taken, in real time (if that double usage didn't confuse you) to travel a set distance. And measuring against cell location or gps location triangulation, it can ascertain instantly, in real time, if there is a problem, with lots of users taking longer than it should for them to traverse a certain route.
This is a two way data path, so the results can also in real time, be transmitted back as RELIABLE traffic data, for sat-nav users receiving it.
Again, this sounds wonderful, whilst the reality will likely not be as good as this at first.
But nevertheless, BOTH of these new approaches are (pardon the pun) driving towards the reliable, useful, sensible level of sat-nav 'intelligence' and worth, that your article hints at.
And whilst it is TomTom that are spearheading these two new approaches, it is not entirely unique, other companies are looking at similar things, and as far as I know there is no patent as such, on the general idea behind them.
So eventually we should see this sort of approach becoming the norm for all sat-navs one day.
But to me, FOR NOW Sat-navs are a GREAT tool to get me to somewhere I don't know how to reach, or where it is located.
But they have never been a tool for getting me there the QUICKEST way, as invariably, local knowledge will always produce a better local route. A sat-nav is just an ideal tool for if you don't even know how to get to a place.
In time, with the sort of things outlined above though, this might actually change!
This is a problem that will be solved with data sharing.
Applications like NM2.0 will eventually permit route feedback to the servers and a learning process starts. The 3G-enabled devices with centralised servers will be the leading navigation devices and Nokia is in a fantastic position to be able to enable this. Many people won't opt-in but there will be enough that do to drive the system into a state where it has far more knowledge about current road conditions than any AA/RAC road traffic service.
S.
Edit - Shadamehr, above, says much the same thing.
I have to say I've had even worse routes offered by Nokia Maps. A trip from the Docklands to Wimbledon send me on a 2 hour journey using a london rign road (not that much traffic) when a trip straight through the center of London took me 45 minutes (on the way back).
This was on a Sunday evening when I should have had no problems. I'm quite sure the calculated route was longer.
A recent comparison of the N95 sitting next to a Tom Tom ended up with the N95 taking me south when I was supposed to be heading north! The journey was a simple A4 to M25 to M40 but the nokia just couldn't get the right route, even after a few recalculations.
Is Tom Tom compatible with the N95's build in GPS yet??
I think that if all in-car gps systems could communicate with each other via a common server then all of these problems would be solved. I'm told that certain modern cars have these systems but I don't know for certain. It would make a lot of sense. Let me give a example of what I mean:
I need to get from the deep, dark suburbs into the city. Normally Route A would be the shortest/fastest choice, but today, due to some incident along the way, all gps's currently travelling Route A, while constantly reporting their speed and position to the common server, show that traffic is only moving at an average of 15km/h. Route D on the other hand, while normally much longer/slower, today is moving at an average of 45km/hour and so is fed to my gps via the common server as a quicker, more fuel efficient and better all around route. My gps then automatically re-routes me along Route D.
As more and more people start to use gps to commute so this system would become more and more efficient. Traffic would be more regulated, less fuel would be used as a whole by the general population. Police traffic departments would love it as their jobs became easier and we would need to rely less and less on radio traffic updates. We would see the end of the massive back-ups suffered the world over and I personally would be able to get the extra 15 minutes sleep in the morning that I so long for. Seems beautiful to me! I cant imagine that it would take too much effort to implement either. Any government with half a brain should be putting this together now, today!
Edit--Woops...Had a few comments come in while I was typing this. Hope I've not been too repetitive!
This one's received quite a bit of coverage lately, it seems pretty similar to the TomTom IQ thing mentioned above, but with live instead of historical info.
Not for me, even if it was UK. High inital cost - over £200 - and a yearly subscription of over a fiver a month.
A lot of people dont like the idea of paying for nokia maps, even though its model is probably cheaper for most in the long run.
It's not the best solution since the maps are not in the device (hence - it uses internet connection), but it have live traffic data taken into account.
http://global.telmap.com/Content.aspx?Page=Live%20Traffic
During navigation, if consideration for traffic is activated (a setting configurable by the user), then Telmap Navigator will always give the optimal route that considers the current traffic situation, whether the user has requested the fastest route or the shortest route.
Thats no different to any other satnav that uses traffic data (Garmin, copilot, etc).
At the moment I'll settle for a satnav that gets me there without taking a silly route. Traffic refinements can wait.
I have to say that I cant believe Steve's route was as bad as that without it being some sort of bug. Especially it being the maps 2 (still beta?) software.
Things like that are exactly the reason I am avoiding it for my reviews until they have sorted it out.
Take Route 66 (version 7) for example. You can set the speeds you generally drive on different kinds of roads. You can also tell it to reduce your overall speed in a city or town by a percentage. So if you tell it that in a city you will only be be to drive at 25% percent of the normal speed, the routing algorithm should take route you around a city if if finds a faster route.
It might be worthwhile to check if your navigation app has something similar, and play a bit with these parameters.
@ svdwal...
Alas, your suggestion is again, like one of the previous posts, not really a solution.
As this would indeed add to the estimated journey time using city centres, as is oft the correct case.
But as "unregistered" higher up says, he had the opposite of this, with a long winded route back home, in order to avoid London driving, when given the time of night on a Sunday evening, a 45 minute direct trip through London would have been much quicker. Your suggestion of adding time to city driving would only have made it worse for him.
In other words, you kindly highlight my point.
It is not just the "where" that the sat-nav's can't get a feel for - the problem is also the "WHEN" factor too.
A road is not simply a road, end of story...
It can be busy at peak times, quiet on weekend evenings, and closed altogether if there is an accident.
So a simple algorithm that says "this road has a 50 mph limit", use that as the criteria for calculating journey times, is realistically, no use. But that's how most current sat-nav's actually work.
What is needed, is the more intelligence led approach, as I mention higher up, where REAL world speeds for actual roads, for different days and times of the day, are KNOWN to the device.
Couple this with the newer REAL TIME Traffic info I mention (not the bog standard usual suspects data we currently receive), and you are a lot of the way there towards having a real, viable, and usable sat-nav solution that Steve alludes to in his article.
I just argue that we should be a lot more closer to this now, than we actually seem to be.
I was always a fan of TomTom since it seemed to make the best job of calculating the best routes.
Since the N95 (and now with my excellent N82) though I have favored those systems which make use of the built in GPS - f which there are a few (Navigon, Nokia Maps, Route 66, Wayfinder).
I have to say that for accuracy and sheer routing performance I have found Navigon the best by far.
It's just a shame that interface is so ugly and you have no option to customise POI's or landmarks.
These days I run my N82 with Nokia Maps 2.0 beta and Navigon.
@shadamehr
Route 66 does in fact offer realtime download of traffic jam information and it is smart enough to reroute you based on the latest info. Biggest problem is (in the Netherlands at least) that traffic jam info is only available for the highways.
There's another problem, which is cost. Getting reliable realtime congestion info isn't going to be cheap. If the savings due to faster trip times are significant, expect to pay as much in total for the congestion info instead. If there are hardly any savings, nobody is going to bother.
There's a very big but to all of this.
In order to build this "real world actual average database", we are going to have to start allowing people to know where we are, where we're going and how fast.
And in today's climate I'm not sure that's a good thing.
Hi
is it me or anyone experience the delays since the new firmware on the N95, i had gps signals in five minutes but since the last firmware upgrade it take more than ten minutes, today going to work at a remote place it took more than thirty minute before i gaveup any
Paul
In order to build this "real world actual average database", we are going to have to start allowing people to know where we are, where we're going and how fast.
Not necessarily, it could be done with samples from volunteers, as is done with television ratings. You don't have to tell anyone which TV programme you watch, but we still know roughly how many people do watch each programme.
@ svdwal...
You miss the point made TWICE now mate, as another user tries to clarify...
It is NOT real-time traffic info - that's the whole point.
You RECEIVE the info in real time, in the sense that it gets to you as quick as possible. It is NOT however, GENERATED in real time. It is standard usual suspects Traffic data, from ICTIS, AA, RAC etc etc, and there is both a delay in them getting it often, and indeed when the issue is cleared, a delay in you being told the route is ok now.
This is bog-standard Traffic, unreliable, oft incorrect, and not trust-worthy, and nothing at all what my original report related to.
You also allude to cost... but that too is incorrect, in that whilst TT are going to charge a premium for this new HD Traffic, that is in order to ALSO actually cover the ongoing Data costs.
Additionally, whilst we do need those transmitting the data, and thus this is Vodafone only at the moment as the sign-up partners, realistically, ANY of us with a data connection can in future both be the receiver of this data, as well as the producer - that is, I already use standard TomTom Traffic - my device linked to my phone via Bluetooth, and connecting using my flat-rate data tariff T-Mobile Web n Walk. I can both be measured to see how long it takes me to travel from one cell to the next, as well as then UPLOAD this info for the benefit of others in terms of allowing this data to be used to measure real time traffic flow (it is submitted anonymously, rest assured).
There is actually no difference in terms of net cost to how I already do things RIGHT NOW.
It is just TomTom who for now are charging a monthly fee for this new service, but only because it covers the SIM card, and data costs already, thus is a ready solution for those not already owning a mobile phone and flat rate data tariff.
So getting back to the point, it is not real time traffic what you mention mate. What I mentioned is the makings of real time traffic though, if it works properly when finally introduced. Hope that clarifies this time.
@ Terry1100.
Not at all mate - not in the slightest - as the data is ANONYMOUS data simply stating which route was traversed, and how long it ACTUALLY took to get from point A to point B, as opposed to how long the map data THOUGHT it should have took.
Provided the data sent is totally anonymous once uploaded, or transmitted, then it can be the most detailed and in depth discourse going for all I care - PROVIDED the details of WHO it relates to is totally gone from it. I've always said I have no problem with the world and his dog knowing every little detail I get up to - PROVIDED they don't actually know it is ME getting up to it, if you follow.
And that is EXACTLY what TomTom did for building up it's IQ Routes data.
There is ZERO need for any company to know WHO the data relates to - all that is needed is to know how LONG it takes to travel a certain road at a certain time. And TomTom's collecting of data in order to provide IQ Routes has always been COMPLETELY anonymous.
Microsoft has invented a system that analysed old traffic data to predict future traffic jams.
http://www.autoblog.com/2008/04/14/microsoft-launches-new-clearflow-traffic-avoidance-system/
"The point of Clearflow is to provide accurate route information that gives you the best chance to avoid traffic on highways and on the side streets."
"is it me or anyone experience the delays since the new firmware on the N95, i had gps signals in five minutes but since the last firmware upgrade it take more than ten minutes, today going to work at a remote place it took more than thirty minute before i gaveup"
Does http://www.allaboutsymbian.com/support/item/Why_is_my_GPS_lock_so_slow.php help?
A recent news article (which as ever I can't find :frown😊 talked about the problems of satnav systems being over optimistic as to the time it would take for various routes and how this through up problems - c.f. if you are not careful, basic satnav systems will take you through a town rather than round the bypass, or will follow narrow backroads, because the fact you can go faster on the bypass is ignored.
I think virtually all the tools suffer from this to a point. The old Nokia maps once directly me off a motorway and then back on at the same junction - seemingly because this was a slightly shorter route than the motorway itself, which went over a bridge (at least I assume that is what it was doing). I've noticed estimates on most of the programs that assume you can average 30mph in central London - technically impossible even at night unless you speed, but definitely impossible in the day. Most of the programs' problems seem to come back to this - they just are wildly out in real delays. I'd rather have something that assumes typical daytime speeds.
Even tomtom has this problem to a point, but Maps is definitely earlier in the learning curve, as it were. However, if you use it out of town, with low traffic, the current version I think is fairly good.
BTW If you are looking for a satnav program for occasional usage, I'd recommend Webraska (www.webraska.com). It is server based, and gives reasonable routes - even so it too suffers from the optimistic problem above. All the routes are downloaded from a server, but this is done in a relatively optimised way - you don't get good maps, but the simplification actually helps in driving😊 Apart from this, its USP is that you can pay by the day (~ �22.50 for 10 separate days usage), which covers the whole of europe, incl. traffic info where they do it. The downside is not so much paying for the data, in my experience, as that when you go off route its natural tendency is to get you back to the original route and not a recalculation from where you are.
@Steve: SatNav is not a system to prevent traffic congestion (yet?). That's the goverments task. However the problem with satnav is mostly inadequate data. Inaccurate data and information will yield improper results. "The map is not the land." I think A.E. van Vogt wrote that first. Adequate information would allow to calculate a mathematical accurate solution. However that might not be enough, talking traffic congestion we also need some form of prediction. Imagine all satnav's deciding to take the same byroads because there is congesting on the A??. I have seen it happen.
The only solution would be to make satnav a traffic regulation tool and set restrictions on your liberties and privacy. Then traffic congestion can be measured, planned and predicted (barring accidents). Needing you to submit a driving plan before take off. Being tracked from A to B. Getting penalted for deviating from the plan and unscheduled breaks...
For now I am very happy with satnav that knows how to get me where I have never been before and knows the presence of roadblocks. Besides 6 out of 10 people still get lost with a satnav...
I also missed a point of view from your 'satnav rant'. Several satnav systems were tested in the Netherlands and judged on their traffic safety. Do they send cars through child-rich urban areas, increasing chances for accidents and increasing local pollution?
Currently satnav are nothing more dan simple calculators and perhaps they should stay like that. If they became as smart as me, they'd allowed to drive to work, without me 😉
snoyt.vox.com
Having tried only route mapping on Nokia Maps it amazes me that if I map a route from my home (Bedfordshire) to my in-laws (Essex) it maps the same route I always take and this is timed at approx 1 hour (which is about right). Now if I map to my parents (who live approx 2 miles from my in-laws at the other end of the road) it maps me on a complete tour of the Cambridgeshire and Essex countryside and is timed at nearly 2 hours. Now it definatley only takes approx 5 minutes to get from one to the other so I am unsure how the 2 hour route is the fastest.
I have worked out about 4 different routes for back up plans and none to date have taken me 2 hours to negotiate.
This is why I rely on a combination of road maps, common sense and route planning to get me from a-b and have not jumped on the Sat Nav band wagon (Mobile Phone or Dedicated unit).
I think it serves you right using BETA software.Nokia Maps 2.0 beta is not reliable.All the forums say that and you have ignored it at your peril.Why is proper Navigation software like Wayfinder and in Version 7 and 8.They have gone through a painful testing and evaluation phase.I was one of the first people to test Wayfinder when it came out properly and gone through many revisions.I use now Wayfinder and Route 66 (because TOMTOM Customer support is useless) only and have the Best user experience!They have never let me down provided I did not switch my brain off,but I would not trust Nokia as they can't even get PC Suite right which is still buggy even after the umpteenth revision.
A new version of wayfinder is being released in May, but I havent been impressed with previous versions.
