More routing oddness

Fanum

Fanum

Toubab
UKGSer Subscriber
Joined
Apr 26, 2003
Messages
48,939
Reaction score
2
Location
Secure unit 2
I set a route to a destination......the Quest gets me there nicely, thanks Betty :thumb

Then I hit and hold the 'find' button to get her to take me home.....and she takes me a different way home??? :confused: :confused:

It gets more confusing though.....last week I went to a site I'd been to a few days earlier.....she took me through the Blackwall tunnel, around the Dalston one way system and up through Islington and Archway.........on the way home, she took me a different way completely from the way she had the previous week......half a mile north of the Dalston one way system, then dropped me back onto the tunnel approach....WTF!!!!!

These were with the same settings, same waypoint, same start point......both routes were good, but they were different :nenau


I think my Betty's got a sense of humour, or at least she likes to vary her route home for a bit of interest! :D

I've also noticed that although a route may well have the same 'pinch points' (like the Dartford tunnel or Blackwall tunnel) , the place you start off from effects the route inbetween two 'pinch points'........even with several miles of shared route at both ends :confused:

I'm sure there are reasons for it, it doesn't bother me 'cos all her routes are valid, but it doesn't make any sense to me :nenau
 
Would it have anything to do with the GPS unit 'learning' your average speeds for certain roads and adjusting accordingly?

I remember PanEuro mentioning this but can't recall if it only applied to the 27xx series or the Quest as well.

Adam
 
Hmm possibly...though the London routes in particular are just city slogs with very similar speeds.......


Actually, no, probably not.......I wouldn't have thought that she'd have 'seen' my speed on one day and decided to route me a different way home the following week based on that...it doesn't make sense :nenau
 
Would it be because, when you hit, Find 'Home', it does not include the waypoints and settings you have inputted, for the way out. It then reverts to whatever you have, 'shortist by distance', or 'time'. Would you not be better to select 'Track back'(can't remember if that's exactly what it's called) option, if you wanted to go the same route home.

This answer is probably a load of gonads cause it's late and I haven't read your thread properly. :thumb
 
Do they really learn, remember and adapt to average speeds on particular roads (or even particular types of road)?

My first real experience of using my Quest 2 came in Spain a couple of weeks ago. Takes a lot of faith to follow instructions when common sense and maps indicate a different route. But on the couple of occasions when she took me down minor roads when the map suggested staying on the main road, I got to the destination within minutes either way of my non-GPS'd friends (either just before or just after) but I'd had the advantage of seeing some much better scenery on twistier roads!

Only one major problem - a closed bridge in Alcaniz nearly made smoke come out. At one point, the instructions in my ear were "Make a U-turn, then make a U-turn!" Had to resort to using road signs - shock, horror - until she sorted herself out a few miles down the road.

In her defence, though, she did have all the tiny little roads in rural Spain in her map. I wasn't convinced before setting off that she'd know them all.

Robin.
 
Schwarz Baron said:
Would it be because, when you hit, Find 'Home', it does not include the waypoints and settings you have inputted, for the way out. It then reverts to whatever you have, 'shortist by distance', or 'time'. Would you not be better to select 'Track back'(can't remember if that's exactly what it's called) option, if you wanted to go the same route home.
Click to expand...

Quite possibly.....there's obviously something different about the way she recalculates the way home using the find button, but the trip to the destination was direct to a waypoint, not via anywhere, and the trip back the same, although I suppose it started me from where i actually was rather than from the exact w/p...that may well have been 100 yrds out :nenau

I'm going to try setting my 'home' point to an exact point...parked on the grating outside my house, and try the same thing again to see what happens.



Schwarz Baron said:
This answer is probably a load of gonads cause it's late and I haven't read your thread properly. :thumb
Click to expand...

Also quite possibly :D :dabone
 
As you travel the Quest is constantly altering the average speed for road types that it uses for estimating time of arrival and for route choice.
 
The distance of a return journey can be different due to junctions and one way streets. This can effect the choice of route it takes.
 
Whatton said:
As you travel the Quest is constantly altering the average speed for road types that it uses for estimating time of arrival and for route choice.
Click to expand...

Well, that's something I've learned today. I thought the estimated time of arrival just changed to account for how much I'd varied from "standard" average speeds on the journey so far. Clever these computer thingies, aren't they?
 
Whatton said:
The distance of a return journey can be different due to junctions and one way streets. This can effect the choice of route it takes.
Click to expand...


Yep...but the first time she took me back round the Dalston one way system.....the next time, she avoided it altogether but the settings were the same :nenau
 
I think it's probably to do with the decision tree and the way it's pruned in order to speed up route planning (I'm making assumptions for how the route finding algorithms work). I'd bore the shit out of you if I went into it but I think the reason you get a different route on the return journey is because it's recalulated the route. But, due to pruning which is necessary because to calculate every single possible route from A to B would take too long/take up too much memory, it uses a heuristic to make an "educated guess" at each stage of the calculation. So, on the return journey it may have pruned out the way you originally came because based on it's knowledge relative to point B it thought some other routes looked better.

That's not at all clear but it's the best I can manage right now.
 
Fanum said:
Yep...but the first time she took me back round the Dalston one way system.....the next time, she avoided it altogether but the settings were the same :nenau
Click to expand...


As you travel the Quest is constantly altering the average speed for road types that it uses for estimating time of arrival and for route choice.
Click to expand...

:thumb
 
unhinged said:
I think it's probably to do with the decision tree and the way it's pruned in order to speed up route planning (I'm making assumptions for how the route finding algorithms work). I'd bore the shit out of you if I went into it but I think the reason you get a different route on the return journey is because it's recalulated the route. But, due to pruning which is necessary because to calculate every single possible route from A to B would take too long/take up too much memory, it uses a heuristic to make an "educated guess" at each stage of the calculation. So, on the return journey it may have pruned out the way you originally came because based on it's knowledge relative to point B it thought some other routes looked better.

That's not at all clear but it's the best I can manage right now.
Click to expand...


Yeah, makes sense to me. Either that, or it re-routed you to stop you having to drive with the sun in your eyes for too long.
 
unhinged said:
I think it's probably to do with the decision tree and the way it's pruned in order to speed up route planning (I'm making assumptions for how the route finding algorithms work). I'd bore the shit out of you if I went into it but I think the reason you get a different route on the return journey is because it's recalulated the route. But, due to pruning which is necessary because to calculate every single possible route from A to B would take too long/take up too much memory, it uses a heuristic to make an "educated guess" at each stage of the calculation. So, on the return journey it may have pruned out the way you originally came because based on it's knowledge relative to point B it thought some other routes looked better.

That's not at all clear but it's the best I can manage right now.
Click to expand...

I can understand why it'd give me a different route on the way home than the rout TO the location.......given one way systems making distances different, it may well tip a marginal decision a different way.....

Why would/could it send me a different way home using the same start and finish points, the same 'hold find button' method and the same settings though ???
 
Fanum said:
Why would/could it send me a different way home using the same start and finish points, the same 'hold find button' method and the same settings though ???
Click to expand...

Chaos theory ?
maybe if there are many possible routes; there will be many borderline decisions to be made at different road junctions; there will be random variations in the performance of the processor (different temperature, different recent use, cosmic radiation etc etc), so you'll get a slightly different choice sometimes, with the same input. In other words, tiny changes in the processor's performance could give a completely different route. I think the idea of different average speeds is probably more important though.

In fact, I wish the garmins gps's did have this function (I believe it exists on tom toms): "recalculate and give me something different". You could cycle through several options and choose the one that looked best.
 
gasman said:
Chaos theory ?
maybe if there are many possible routes; there will be many borderline decisions to be made at different road junctions; there will be random variations in the performance of the processor (different temperature, different recent use, cosmic radiation etc etc), so you'll get a slightly different choice sometimes, with the same input. In other words, tiny changes in the processor's performance could give a completely different route. I think the idea of different average speeds is probably more important though.

In fact, I wish the garmins gps's did have this function (I believe it exists on tom toms): "recalculate and give me something different". You could cycle through several options and choose the one that looked best.
Click to expand...

See post #15 ;)
 
Fanum said:
....last week I went to a site I'd been to a few days earlier.....she took me through the Blackwall tunnel, around the Dalston one way system and up through Islington and Archway.........on the way home, she took me a different way completely from the way she had the previous week......half a mile north of the Dalston one way system, then dropped me back onto the tunnel approach....WTF!!!!!

These were with the same settings, same waypoint, same start point......both routes were good, but they were different...
Click to expand...

Hi Fanum:

The reason for the different routes is probably simpler than anyone has thought so far: Different times of day.

As the cartographic database becomes more detailed and sophisticated, time of day restrictions get added. For example, you might not be permitted to make turns at certain intersections at different times of day. Or, a road that might be three lanes wide during most of a 24 hour day might be restricted due to the presence of a bus or car pool lane during rush hours. Or, due to no parking and no stopping restrictions during rush hour, a road might be (effectively) wider during rush hour than it is during the middle of the day.

Lastly, traffic light patterns might be synchronized along a certain stretch of road during rush hours, thus increasing the average speed that can be achieved along this road during periods of heavy traffic flow.

What we see on the map screen is only a fraction - a tiny tip of the iceberg - of the data that is contained in the cartographic database. Each and every road has 'street attributes' attached to it. There can be as many as eighty attributes per roadway, and we don't see any of them - but all of them are considered by the routing engine in the GPSR when we are out there on the road.

If you calculate a route across Central London using MapSource, then you download it to your GPSR, chances are that you will soon see a message saying "A better route is available" (if you have a fairly new GPSR, that is). You are seeing that message because the GPSR has had a second look at the route in the context of the actual time of day (and, frequently, day of the week!) that you are trying to navigate it, and has discovered time dependent road attributes that were not considered either the first time you created it on the GPSR (at a different time of day), or when you created it on MapSource (which doesn't ask you what time of day you intend to drive the route you are creating).

Michael
 
Hmmm...makes sense Michael, but I don't think there are any of these varients present in the small sections that Betty routed me around......and there certainly aren't any that are dependant on time of day etc on the countryside routes across Kent that she sent me different ways home on :nenau

Mine's a Quest.....and I'm still using ECS 6.......so it's not that sophisticated either. ;)

Like I said, I'm not particulalry bothered about it, it's just odd, and rather irrational on the face of it :)

Thanks though :thumb
 
Fanum said:
Mine's a Quest.....and I'm still using ECS 6.......so it's not that sophisticated either...
Click to expand...

It may not have been that sophisticated when you first took it out of the box, but have you updated the system software (the software inside the GPSR) since you bought it? It is possible that the system software is considering more variables - more road attributes - than it considered before. In other words, some of those time of day related variables may have been on your cartography all along, but the routing engine has been improved so that it now takes them into consideration when a route is calculated.

I don't know this to be a fact, I'm just mentioning it as a possibility.

Michael
 
Nippy Normans supply Touratech, Wunderlich, Desert Fox and many other makes of accessories.
You must log in or register to reply here.

Similar threads

stick
Stopping re routing
Replies
2
Views
214
Andyshaggy
Wapping
  • Sticky
Using MRARouteplanner created routes on the BMW ConnectedRide Navigator
Replies
2
Views
933
Wapping
Wapping
Mzokk
Review of Tourstart
Replies
4
Views
488
Mzokk
Mzokk
France and Spain June 2022
Replies
3
Views
669
leedude03
leedude03
How to use MyRoute to help having ‘starbursts’ of riders, all apparently using the ‘same’ route
Replies
0
Views
538
Rasher


Back
Top Bottom