Traffic Lights and Stop Signs in Route Calculations

 

I haven't been able to locate anything on this topic in these forums, but if I did miss a thread, I'd be grateful if someone could point me to it.

It didn't take long to become apparent to me that when calculating routes, my 760 doesn't factor in delays at stop signs and traffic signals. With the calculation option set for fastest time, results will routinely favor shorter routes that have more stops and lights that actually take longer to navigate.

Is Garmin (or any other GPS manufacturer, for that matter) doing any work on incorporating this type of data into their routing algorithms? I don't pretend to know anything about that kind of coding, but it doesn't strike me as anything that would be hugely difficult, considering everything else they've been able to accomplish. Collecting the light and stop-sign location data might even be the bigger effort.

Jes' wonderin', is all...

faster/shorter

Can't say how things are calculated, but from my experience using Faster/Shorter routing...

In my area of travel, Shorter usually results in a route through neighborhoods, usually with more stop signs etc. Faster usually routes me around these neighborhoods in favor of more major highways/interstates, resulting in a Faster eta. So, in my area of frequent travel, Faster usually results in a faster arrival time. Shorter may result in less miles but a longer eta.

--
........Garmin StreetPilot c550 / Nüvi 765...........

Re: Route Calculations

Felix Krull wrote:

Is Garmin (or any other GPS manufacturer, for that matter) doing any work on incorporating this type of data into their routing algorithms?

Check out TomTom models with IQ Routes. This technology calculates your routes and ETAs based on the real average speed measured on roads rather than speed limits.

Krieger

Garmins also learn from your driving habits

Krieger wrote:
Felix Krull wrote:

Is Garmin (or any other GPS manufacturer, for that matter) doing any work on incorporating this type of data into their routing algorithms?

Check out TomTom models with IQ Routes. This technology calculates your routes and ETAs based on the real average speed measured on roads rather than speed limits.

Krieger

The thread link below is about the Garmin units learning based on your driving habits on the 10 road types:

http://www.poi-factory.com/node/17641

--
Garmin Drive Smart 55 - Samsung Note 10 Smartphone with Google Maps & HERE Apps

255W set for 'Faster'

I have my Nuvi 255W set for 'Faster' and it routinely tries to route me through surface streets even though the freeway is MUCH faster and I live less than a mile from a freeway on ramp....I'm still trying to fugure out the logic on that one.

Is there anything from GRMN or a 3rd party on this?

rjrsw wrote:

The thread link below is about the Garmin units learning based on your driving habits on the 10 road types:

http://www.poi-factory.com/node/17641

I looked at that thread, but nobody gave any references or links to substantiate the existence of adaptive behavior on their products. This seems rather subjective, as there were also some users perceived no change over time.

Garmin makes no claim of any adaptive behavior on their website either.

If such technology indeed exists on Garmins, what is seems to do is rather track preferred segments of traveled routes and give them priority, regardless of being part of the optimum route or not.

Krieger

i totally understand where

i totally understand where you coming from, as dicussed in the other post, allegedly, the average speed of 10 different types of roads....i would guess eventually as you drive, and stop on these types of roads the gps should use that to lower the avg speed on that type of road, but i dont know if it calculates the time stopped, or use that for the average.

Anytime i have local driving the ETA is severely off.

Comparison between TomTom and Garmin arrival time estimating

Krieger wrote:
rjrsw wrote:

The thread link below is about the Garmin units learning based on your driving habits on the 10 road types:

http://www.poi-factory.com/node/17641

I looked at that thread, but nobody gave any references or links to substantiate the existence of adaptive behavior on their products. This seems rather subjective, as there were also some users perceived no change over time.

Garmin makes no claim of any adaptive behavior on their website either.

If such technology indeed exists on Garmins, what is seems to do is rather track preferred segments of traveled routes and give them priority, regardless of being part of the optimum route or not.

Krieger

I noticed it myself and read on one of the GPS 7XX review sites that they also noticed that the Garmin was adjusting it's 10 road type speeds over time. I noticed that it was getting more and more acurate on the original arrival estimate and the actual arrival time.

I verified it by comparing the early route simulation speeds I had seen on a certain road type which were the normal posted speed to the route simulation speed now and noticed it increased by 4-5 mph over the initial and that is probably more with my actual driving speed.

Another thread link comparing TomTom and Garmin's estimated arrival times:

http://www.gpspassion.com/forumsen/topic.asp?TOPIC_ID=104767

Another discussion thread on this subject:

http://www.poi-factory.com/node/12011

--
Garmin Drive Smart 55 - Samsung Note 10 Smartphone with Google Maps & HERE Apps

Avoidance?

thomas8093, a dumb question, but are your avoidances set to avoid highways or toll roads, if that's your case? This happened to my father when he first got his GPS.

--
Nuvi260

I did find it curious in a

I did find it curious in a simulation that even though there was a 55 speed limit sign on the left of the screen, my nuvi 750 was doing 65 mph in the simulation. If we have adaptive electonic thermostats, why not adaptive GPS? I have only had it for a week, but did do some spirited highway driving.

I did tend to take different routes than suggested to avoid areas I knew had many stop lights and possible traffic. So I changed Australian Karen to say "Your every wish is my command, Dave" instead of "recalculating".

--
nuvi 750, eTrex Legend HCx, Mobile 10/Palm TX, GPS 45

Re: Comparison between TomTom and Garmin arrival time estimating

rjrsw wrote:

I noticed it myself and read on one of the GPS 7XX review sites that they also noticed that the Garmin was adjusting it's 10 road type speeds over time. I noticed that it was getting more and more acurate on the original arrival estimate and the actual arrival time.

I verified it by comparing the early route simulation speeds I had seen on a certain road type which were the normal posted speed to the route simulation speed now and noticed it increased by 4-5 mph over the initial and that is probably more with my actual driving speed.

Another thread link comparing TomTom and Garmin's estimated arrival times:

http://www.gpspassion.com/forumsen/topic.asp?TOPIC_ID=104767

Another discussion thread on this subject:

http://www.poi-factory.com/node/12011

I understand what you are saying, but we are not talking about the same subject: those two threads you point to discuss ETA calculation (which is adjusted as you go). Again, this is perceived subjectively, as different people in both threads feel different about the accuracy of both manufacturers.

Instead, Felix Krull's question is regarding the routing algorithms (computing the best way to navigate between A and B). Newer TomToms use the data from not just one user, but from all the users that have traveled the roads considered in a route and that have agreed to share this information, to assign a real-world weight to each road. It also takes into consideration the day of the week and time of day. It then compares its original ETA, against the actual time to travel the route.

Garmin does not have anything equivalent to this. It's routing engine only considers the posted speed limit to generate routes.

http://www.gpsreview.net/tomtom-iq-routes/

Krieger.

That is definitely

That is definitely interesting. i do think however if its a possibility of overkill. Are ETA's that off? or that important? I do respect the effort being put into moving forward with technology. How ever that is not enough to make me switch to a tom tom. I think it would definitely take time to create an accurate database, especially for each day, time, for each segment. Not to mention other factors that may effect the average times, such as accidents, bad weather. I think there are too many variables to ever get an accurate reading, and like I said I don't think the ETA's we have now, are that Off. It usually averages out, driving slightly above limit buys you some time you'll kill stopped at lights. This new technology might throw off that balance lol.