IQ Routes multiple questions

[leor]

My commute home hits heavy traffic on a particular bridge everyday. When I leave my office IQ Routing tells me I will be home in 12 minutes, but it usually takes at least 15 and sometimes closer to 20.

Is IQ routing preset with historical data gathered prior to the release of the 930/730, or is it still learning and updating driving times based on new data gathered from current users? Will the accuracy improve over time?

Does it update via mapshare?

Should we always have our TomToms on (even when we don't need them) just to gather data and improve IQ Routing accuracy?

Thanks.

 

[Michael Quinlan]

IQ Routes is indeed based on historical data. However, the data used is a daily average. A slowdown at rush hour is not going to have a significant effect on the daily average.

As for the rest of your questions, I don't know... but I would like to!

 

[leor]

Thanks, Michael.

I found this IQ Routing information on the TomTom website:

------------------
How it works

Millions of TomTom users worldwide voluntarily provide us with anonymous historical speed-data from each of their journeys every time they connect their device to TomTom HOME .

As a result, TomTom now has a huge database, containing billions of miles of real customer driving experience, collected over the years from more than 7 million TomTom users.

Using this data, TomTom?s unique IQ Routes? technology calculates your optimal route. This route is based on actual speed data driven on roads, not just maximum speed limits, taking into account any hold-ups that might slow you down (busy junctions, zebra crossings and even shopping crowds).

This data enhances the quality of TomTom navigation across the board, but is particularly helpful in improving the quality of detailed routing we give you for the specific time of day and day of the week of your chosen trip.

-------------------------

They do say "specific time of day and day of week" so I think it should be taking into account rush hour traffic, and not just give a daily average for a particular route. However, as you suggest, it appears to just use a daily average so I guess TomTom's website claim is false. Shame on TomTom for advertising something they do not appear to be delivering.

For what it's worth, I am keeping my TomTom on during my commute both ways every day (I use it for my bluetooth phone anyway) so I hope my daily data is being logged somewhere. Maybe IQ routing will be improved or enhanced in the future.

 

[Michael Quinlan]

While it does say that routing is for a specific time of day and day of week, I haven't seen any differences in routing based on either (but routing *IS* sometimes different if I turn of IQ Routes). A telltale sign that time of day is not considered (yet) is that when planning a route (providing a starting point and destination) you are asked to specify a day of week, but not a time of day.

While I can't factually dispute what they say, I do see evidence that contradicts their claim.

 

[spud]

My IQ routing changes the route for the same trip between weekdays and weekends.

 

[Buddharay]

it's all about the average speeds (instead of the speed limit). i think it depends from case to case, but in theory it should work pretty well.

 

[leor]

My IQ routing changes the route for the same trip between weekdays and weekends.

Spud, I'm in Vancouver too. The bridge in my post is Knight Street. As you know, getting across town in any direction during rush hour is VERY different than the same trip on the same day at 10pm. As it is implemented now, I think IQ Routes will give the same trip time for both trips. I will input a few dummy trips at different times of the day to confirm this, but I think I will find this to be true.

TomTom is ahead of everyone else with this technology and it is a great start but only moderately useful right now. You still have to guess as to how much time to add to a trip if you have to get to an appointment at a certain time. Let's hope TomTom further develops the technology to make it really useful. The data is there (or is being collected on a continuous basis), they just have to use it.

 

[Michael Quinlan]

Spud, I'm in Vancouver too. The bridge in my post is Knight Street. As you know, getting across town in any direction during rush hour is VERY different than the same trip on the same day at 10pm. As it is implemented now, I think IQ Routes will give the same trip time for both trips. I will input a few dummy trips at different times of the day to confirm this, but I think I will find this to be true.

TomTom is ahead of everyone else with this technology and it is a great start but only moderately useful right now. You still have to guess as to how much time to add to a trip if you have to get to an appointment at a certain time. Let's hope TomTom further develops the technology to make it really useful. The data is there (or is being collected on a continuous basis), they just have to use it.

The data they are collecting obviously has some sort of date/time associated with it. I think the biggest obstacles to including time of day in the calculations would be the CPU's processing speed, and the possibility that the data required for time of day calculations could have a significant effect on map size - possibly making the maps too large for current devices. Even if they just included hourly averages it would mean a 24-fold increase in the size of the data. I don't see any benefit in going beyond a 10-15 minute granularity, but that would mean a 100-150-fold increase in data size (from current daily averages).