Routing Algorithms

[rsforkner]

I am really curious if there is a place I can go to understand the routing algorithms used in the TT One. It seems like the TT always used the route with the highest speed limit for the fastest route regardless of other factors.

For example: A store I visit about 3 miles from home I use the TT to find the ?fastest? route home. Route One (chosen by TT) the speed limit is 45 mph and contains 8 traffic lights. My preferred route the speed limit is 40 mph but only has 3 traffic lights. The second route is fastest 90% of the time because of the fewer lights. The difference in distance is about 1/10 of a mile.

If I head for the second route the TT wants me to turn around and go back to the first route until the distance is almost a mile longer before it gives up and plots the second route.

I really would like to understand exactly what is going on here.
Thanks,
Bob

 

[Arno]

I don't know the TT One but would guess that higher end phones, capable of IQ Routes would take that into consideration.
Do you have the option to Avoid part of route and then select the road you do not want?
That may have to be done more than once to get it exactly how you want it but, even then it is a one off deal unless you make that route an itinerary and select it the next time you want to go there.

I doubt very much that TomTom would make their intellectual property public and for all competitors to see.

 

[rsforkner]

Thanks for the reply. It is mostly a matter of curiosity at this point. Last summer while on a long trip in unfamiliar territory I figured out that about 20% of the time finding an address is mostly an "estimate". The other 80% of the time it will get you within sight of your destination. Now I am trying to look at other attributes of the GPS system.

Around home, of course I really don't need or rely on it other than curiosity. I wasn't sure if the antilogarithm would be considered intellectual property. I wouldn't be surprised it they were.

Of course, both routes get me to my destination.

I did choose "fastest" route and did not exclude any road types.

Thanks,
Bob

 

[Arno]

Last summer while on a long trip in unfamiliar territory I figured out that about 20% of the time finding an address is mostly an "estimate". The other 80% of the time it will get you within sight of your destination.

Well, in the 22 days and 14,000 km this summer Kate got me to every single motel, attraction and point of interest without fail.

Around home, of course I really don't need or rely on it other than curiosity.

I had drifted out of this list before my trip and hadn't been using the GPS since that trip when I convinced a computer phobic friend of mine to buy a 930.
With him asking all kinds of questions I poked my nose in here again.
As of October 26 it has become unlawful to use a hand held communication device while driving. Since then I use the TomTom even on short trips and use these opportunities to test some of the matters discussed here.

I did choose "fastest" route and did not exclude any road types.

Well your post reminded me of similarities. Unless away SWMBO and I go at lunch time to a shopping center for coffees. Kate wants me to drive a faster speed limit road with eight lights but I prefer an other with four lights, one of which has a very long left turn priority phase which I must take.
Took a stop watch with me and, bloody 'ell, her route did take a minute less then mine.

 

[rsforkner]

Last August I traveled about 4300 miles (6900 km) in 21 days. Most of the time it was dead on or within a couple of buildings of where I wanted to go. In one case it was over .5 miles away but I consider that a rare exception. A couple of other times it was 2-3 blocks away but I still found my destination. (My own house is off by 3 blocks) I always tell folks using GPS to find me that when their GPS says they are there, they are really 3 blocks south of my house. <smile>

I have a Parrot hands free device for phone calls so I don't really need one in the TT for calls.

In my little trips around home the time is always within a minute or two looking at my watch. I suppose at certain times of day and year the "preferred" route may be a bit faster when there is very little traffic. Art those times I may be able to sail through most of the lights.

I am always curious about how these kinds of things work. (It drives my wife nuts)

Bob

 

[mvl]

If I head for the second route the TT wants me to turn around and go back to the first route until the distance is almost a mile longer before it gives up and plots the second route.

From your earlier posts it seems you have the One 125. That model does not have IQroutes, so it doesn't take into account real-world driving history, and thus doesn't take into account things like traffic lights.

Instead, the One 125 divides roads in to a couple of road classes. eg: highways, main streets, sidestreets, etc. Each class is assigned an assumed speed (try running a route demo to see how fast the assigned speed is). Then Tomtom tries to find the fastest route using those assigned speeds.

The assignment is pretty poor, which is why Tomtom switched to an IQroutes real-world algorithm in the newer models.

My guess is that your road that has lots of stoplights is assigned a faster "class". Sometimes this happens because the road has a state highway number assigned to it, sometimes this happens if the road is physically wider, other times you have no idea why Tomtom chose this higher class.

 

[rsforkner]

THANK YOU

That is exactly the kind of information I was looking for. Understanding this information makes me feel better about the limitations of the One 125.

Considering the price point I can accept the the limitations of the unit.

You are correct regarding the "type/size" of the roads. Route one is six lanes and the preferred route is four lanes.

Again, thank you,

Bob