How iQue figures driving speed

[jamminrock2]

Being one of those people that has to click on everything there is, I THINK I found out how the iQue figures out the driving speeds.

If you open the mapsource program, not palm desktop that comes with the iQue, but the program you have to load from the Garmin site, click edit, preferences, and then the routing tab.

Lo and behold, there is a section called driving speeds, where I think you can see the Garmin default speed. In mapsource you can enter your own average driving speeds, but I'm not sure the iQue will accept this. I do however assume that this might be the Garmin default?

in MPH

interstate highways = 67
major highways = 58
other highways = 45
collector roads = 35
residential streets = 25

I hope this will help all of those that have wondered about this. I know I've seen a few threads on this subject. I'll attempt to test this, and see if the iQue will accept a change made to driving speeds. From my last trip, it seems that these might really be the driving speeds used though.

[apersson850]

Yes, it's likely that they picked the same speeds as the default setting in the iQue.

No, the user can't change these values in the iQue. They are in the NAV Speed Category Data in non-volatile memory. But only at Garmin do they have software to access these databases directly. Yet.

[stephanpls]

The 67 MPH for US interstate highways equals 107.8 km/hr, which matches my observation on a Dutch highway when the iQue calculated a 6:00 driving time for a distance of 10.8 kms.

But..... this morning while driving to my work the iQue calculated with a speed of 98 km/hr for a leg that was only highway, exactly the same type as stated above. It keeps puzzling me.

[apersson850]

I know for sure, that the ETA and ETE values aren't derived 100% from the speed category data, but also influenced to some degree (unsure how much) by how you've been driving lately.
They actually tried to make the route choice adaptive in the past (before the iQue), by taking previous progress along different roads into the calculation, but in the real life, that turned out to be too unstable and erratic to be useful.

Probably too much feedback in that algorithm, so it got unstable to some extent. I mean that the changes that previous progress caused when selecting different routes, may have induced even larger changes in the real progress along the roads, which in turn caused... You get the idea, I suppose.

[Rod Bunker]

Yesterday evening about 430PM I left the West Chicago suburb of Aurora heading about 140 miles to Indianapolis. When the que finished the routing it gave my arrival time as 817PM. I guess 5PM traffic held me up, it missed the drive by 1 minute. I drove into my motel in south Indy at 818PM. The rest room stop could have been the culprit. I am impressed. The routing was exactly as the one I always used.

[stephanpls]

Rod,

I think that only means that you averagely drove according to the speed the iQue expects you to drive.
Last night I drove home and the iQue used the 98 km/hr for highways in its ETA calculation. I got stuck in a terrible jam, and missed the initially calculated ETA by 1/2 hour!

[Atlrodeopup]

So are you saying (forgive my possible ignorance), that if you are out and about then decide to go somewhere, the time it thinks you will arrive etc is based on these pre-figured speed averages and not based on your actual speed as determined by the gps itself?

If not - what did you mean? Sorry for sounding dumb, but maybe I missed the point of the original post.

[stevevo]

Originally posted by stephanpls
Rod,

I think that only means that you averagely drove according to the speed the iQue expects you to drive.
Last night I drove home and the iQue used the 98 km/hr for highways in its ETA calculation. I got stuck in a terrible jam, and missed the initially calculated ETA by 1/2 hour!

I think the iQue has the data to know what the speed of certain roads are and calculate that into its ETA. I have used it many times on the road and find it to be about 90% correct with the prediction of ETA. I do gain some time over the extended drive as I may be driving a little faster than the speed limit. It seems I gain about 1 minute every 15 of driving time but usually will arrive at the destination with in a couple of minutes of the ETA.

[stephanpls]

Yes, the iQue has a sort of typical speed (which is a little slower than the maximum speed) for each type of road. With that speed an ETA can be calculated. If you drive moderately (not too fast or too slowly compared to the typical speed of the road ) and if there are no delays by ljams or any obstacle the initially calculated ETA will be correct within a few minutes.

[apersson850]

Originally posted by Atlrodeopup
So are you saying (forgive my possible ignorance), that if you are out and about then decide to go somewhere, the time it thinks you will arrive etc is based on these pre-figured speed averages and not based on your actual speed as determined by the gps itself? Yes, that's correct. And if you think about it, it's better that way.
If you are driving on the highway, at say 120 km/h, and the last 30 km to your final destination are roads suitable for 45 km/h, then the ETA would be very optimistic, if the iQue didn't check the likely speed over the entire route.

But then it's of course possible to take deviations from the expected speed into account. Like if you consistently have been doing 120 km/h on 100 km/h roads, then there is some logic in assuming that you'll do 60 km/h on 50 km/h roads.
Another day you are perhaps doing only 80 km/h on these 100 km/h road, so maybe 40 km/h could then be expected on the 50 km/h road.

For example.

[wa4phy]

From my observations, and based on others as well, the 67 mph speed that is used, is very close, if not exact to what my speedometer reads when doing 70 mph, which also happens to be the most widely accepted interstate speed, barring CO, TX, NM, AZ, and some of the other western states. I checked the Que speed against radar on 3 instances yesterday driving thru Alabama on I-20, E of Birmingham. The Que was exactly with the radar indicated speed. Neat!

Sam

[L8Apex]

I've found the ETA to be horribly inaccurate except where the route is such that no variations can be introduced from traffic or stoplights. If I so much as have to tap my brakes because of a slow car in front of me or a stoplight, I'll arrive late.

It's hopeless during rush hour. I commute 30 miles between a west surburb of Chicago and the city every day. With no traffic, I can complete this drive in about 35 mins, which is the estimate the iQue gives me. But when traffic is bad, the drive easily takes 3x as long, or 1hr 45mins. And typical travel time for me is 60 mins. So I'm typically late by more than 70% of the iQue's estimated travel time.

The iQue does NOT seem to account for slow speed when I'm stuck in traffic. For instance, let's say my hypothetical route is along 30 miles of highway and the iQue thinks my average speed will be 60mph (just to make calculations easy for this example) so my ETA is 30mins from now. But in reality, the highway is backed up the whole way at an average speed of 20mph. You would think that after the first 10 miles, 30 minutes into my drive, the iQue would realize that I'm averaging 1/3 of the expected speed, and since I still have 20 miles left to go at an average of 20mph, my new ETA from that point should be 1 hour from then. But the iQue appears to simply recalculate my ETA from my current position based on the speed of the road, so it will display an ETA of 20 minutes from then from that point. There's a huge discrepancy between 20 minutes and an hour. Likewise, once I'm 20 miles into my route (after an hour has passed), it will say I have only 10 mins left when I actually have another 30 mins to go. It doesn't catch on.

I end up doing the math in my head as I drive to estimate my ETA myself. I find the iQue's ETA useless in normal every-day driving.

Now on a roadtrip across several states of unfamiliar highway where I'm scared to drive too fast because I don't know the hiding spots of the cops... In that case, the iQue's ETA is almost dead on, since there are no stoplights and no other cars to slow down for.

In Garmin's favor, I'll say that it's actually very difficult to estimate travel time when the route can potentially have a wide variety of delays. Just because I'm stuck at 20mph for the first third of my drive doesn't mean I won't suddenly hit wide-open freeway and average 80mph the rest of the way. It's impossible to predict, and there might not be enough historical data at any given time to make a reasonable estimate.

[Peztone]

This is interesting.. because my iQue does move the ETA based on my moving speed, as an example I left central London at 4pm the other day and it said I'd be home at 5:05, I knew damn well I'd be home at 7pm, and 1 minute at a time the iQue slipped the ETA until about 30 mins from home it was saying 7pm..

Also if I should ever break the speed limit (lord forfend) it says I'll get there earlier..

Are you patched up to date?

[L8Apex]

Peztone-

It might be doing some amount of correction; I'm not sure.

Did slow conditions continue during the last portion of your drive? It's possible that driving conditions changed by the time you were 30 mins from home, such that the rest of your drive wasn't delayed from that point on, making the iQue accurate at that point. This happens to me frequently. The last 15 mins of my commute home is usually past the delay, and roughly accurate ETA from that point on.

I'm not saying the iQue does no correction at all for current speed. Just that I've been unable to verify any such correction, and observed many cases where it does not account for delays. I'm running Release2 firmware with the latest patch. Maybe my typical driving conditions confuse it. *shrug*

[Peztone]

It does sound like yours is similar to mine.

If I'm driving back from central London, the first 60 mins of the drive will be at 3mph so I sit and watch the iQue add 1 min to the ETA for every minute I sit in traffic, then the drive speeds up to 30mph for about 30 mins so the iQue only adds a minute every once in a while (I'm on 50mph roads at this point) then the last 45 mins is at 70-90mph on a Motorway, by this time the iQue has recalculated the eta and it's fairly accurate..

But I do think they should factor in some major city rushour madness.

If I just drive around out here in the countryside where there is no traffic the ETA is always very accurate.

-S.

[reinbeau]

Originally posted by Peztone
This is interesting.. because my iQue does move the ETA based on my moving speed, as an example I left central London at 4pm the other day and it said I'd be home at 5:05, I knew damn well I'd be home at 7pm, and 1 minute at a time the iQue slipped the ETA until about 30 mins from home it was saying 7pm..

Also if I should ever break the speed limit (lord forfend) it says I'll get there earlier..

Are you patched up to date? Yep, mine adjusts the ETA, too. I've been kinda impressed with that, as has hubby.

[RhinoDoc]

Earlier versions of my GPS's did not correct for differing speeds. By the time I got the GPSIII it stopped the ticking clock when I was stopped and adjusted for real speed, in spite of what the default speed limit on the road was. This seems to continue with the iQue.

[microbe]

On a given stretch of freeway 5/405 in southern California that I have been driving a lot, my iQue uses a speed of 74mph to calculate the ETA. This speed is close to the the average traffic (and mine) speed on that road.

On other freeways that I have not driven before with iQue, it does use exactly 67mph to do the ETA calculations.

Now the test to do is to use two iQues at the same time to see if there is any difference in ETA based on past history.

[luserSPAZ]

From my iQue experience, I don't believe the "adjusting ETA" is based on how fast or slow you've been driving, but simply by how much route you have left to travel.

Basically, the iQue calculates the amount of time it thinks your trip will take by taking each road (or section of road) in your route, and dividing by its internal speed limit for that type of road to get the estimated time to drive that stretch of road, then adding them all together. Then, as you travel the route, it simply figures the time left to travel is the time it would take to drive the remaining part of your route.

So, for example, if your route uses 3 roads, and the iQue figures it will take you 10 minutes to drive each road, then it will estimate 30 minutes driving time. If you drive very fast (like me) and finish the first road in 5 minutes, then the iQue will still display 20 minutes as the time remaining in the trip, because it knows you still have those other two roads to go through. This means that if you left at 1:00, it would initially display your ETA as 1:30. After the first road, it would be 1:05, so your ETA would become 1:25. It's not directly taking into account your speedy driving, but it does change the ETA. By the same token, if you drove slower than it expected, it would slip the ETA.

This is all based on my experience with the iQue, so I could be wrong, but it seems to match up with what I've seen. I don't believe it ever changes its route time estimations, it just changes the ETA based on the amount of route remaining.