jime

The trip Estimated Time Remaining (ETR) on my TL Navigation system is very inaccurate. I would estimate about a 25% error rate, making the “TIME TO GO” feature totally useless. It makes me think that the Navigation system may not be working properly.

The ETR function of my Garmin Streetpilot III GPS is always within a minute or so on a one hour trip. For comparison, my TL Nav. system is off about 6 to 8 minutes on a 30 minute trip. The Garmin keeps a historical track history and adjusts the ETR for driving habits and conditions. I would think that the TL Nav. system would have the same capability.

I was wondering if anyone knew where I could get more information about how the TL Nav. system calculates the ETR. If I knew how it was calculated, I could determine if it was working properly. If it is working properly, maybe I could make some mental adjustments for the calculation errors.

Thanks,

vtechbrain

I don't know either but it is not very accurate, although 25% seems a little steep. That might be true for a short trip but for longer over (100 miles) I would guess around 10 percent. It is just a toy anyway since the variables of ETA are many.

Jss1

I believe it uses the posted speed limits when available. When there are no speeds posted I read it used 40km (25mph) as default. I would think the distance would be more accurate.

nickpoore

Yeah, I agree with JSS.

On longer freeway trips, I often see the ETR time adjusting itself because my acutal speed is different to the posted speed.

Simple way to do this is to pull over on any trip and just sit there. The ETR will remain the same (as your distance to destination has remained the same) even though your average speed is slowing down.

I like the ETR function, but I accept the simple fact that computers can't count.

jime

But computers can count and they can adjust the ETR for the individual driver.

As I stated in my first post, my Garmin Streetpilot III Automotive Navigation System is accurate to within 1 minute or so for an hour drive. It accomplishes this by recording the history of previous drives, and estimating the ETR based on historic speeds as compared to posted speed limits.

My Garmin costs about $700. I expect the $2000 system in my TL to also have this feature.

supergreen

last time i was heading out to vegas it seemed like it was using 70mph to judge the time.

Modeler

And you would be wrong! It does not have any capability to collect and analyze historical data.

Other posters are also wrong about what speeds are being used to calculate the ETR. It's only the posted speed limit for those roads for which NavTech has no information. Which is a very small percentage of roads, usually rural.

Majority of roads have what's called a "free-flow" speed attribute in NavTech's dataset. That's the speed that vehicles have been found to travel on this road, when no congestion is present. For those roads that the actual data is not available, they may simply use an average speed that depends on the road type (local residential street, minor collector, major collector, expressway, etc.). In other words, the speeds are more reflective of the actual observed speeds than posted speed limits.

drising

I'll make the issue even more complex.

On a recent trip on both 2-lane state highways (first) then interstate, ETA was first showing 9 hours for a 4 hour trip. This ETA was based on a great deal of stop-and-go traffic through small towns.

Once we were on interstate, the computer fairly quickly reduced the drive time, eventually ending up at the remaining time of what we again knew to be a 4 hour drive.

This leads me to believe the ETA is based on "current" speed (probably within a frame of recent time/mph).

I'll have to check the nav manual for verification.

fuque

Nav systems typically use fixed (but configurable) values for the expected travel speed on each type of road (e.g. 65 on major freeways, 55 on 2-lane country roads, 30 in town, etc.) In other words, the nav system stupidly uses these fixed values, totally ignoring your current speed. Thus, even if you're driving 100 MPH, the system will calcluate your ETA based on the 65 MPH speed value setting for the type of road you're driving on.

This was how the nav system in my IS-300 operated, and it used to WILDLY overestimate the time to destination since most rural Interstates in the west have 75 or 70 MPH speed limits, but it was stupidly set to 60 MPH (and I didn't know those settings existed). I haven't bothered to check to see if the Acura nav works the same way, but that's the first thing you should check if you're getting inaccurate estimates.

HTH!

¿GotJazz?

As with most things on GPS, it's up to the system designer on how they want to implement it. There's many ways to skin this cat.

Note: All of the ways that I'll mention require that the GPS system is aware of speed limits on each road, in some manner.

Option (1): Easiest. The GPS database has all the posted speed limits stored in the database. Assume that you will average the posted speed limit on each road. This is probably the least accurate, based on how I've seen people drive in this country. Everyone drives faster than the speed limit (except my girlfriend, and she wants the dealership to install a speed governor in my new TL).

Option (2): Very slightly more difficult. Instead of having the speed limit for every section of road stored in the database, store what type of road (interstate highway, state highway, neighborhood street, 2-lane highway, etc.) in the database. Then, let the user (you) manually enter what you would tend to average on that type of road. For example, 2-lane highways may be posted as 55 MPH, but you may average 63 MPH on that type of road, so you would enter '63'. Could possibly switch these stored values based on the seat memory selection from your remote (In case your spouse drives like a maniac). More accurate than option (1).

Option (3): More difficult. The GPS database keeps two sets of numbers for each section of road, using a combination of Options (1) and (2). The database would contain the posted speed limt (or road type), and it would also store a variable that contains how fast you drive for that speed limit or road type. This value would be constantly updated based upon your driving habits, being done behind the scenes. More accurate than option (2), and less user interfacing.

Option (4): Most difficult. Like Option (3), but enhanced to keep track of your average speed for every section of road in the country. The base values that the GPS system would use for an undriven section would be the constantly updated values derived in Option (3), which would change once you drive on a previously undriven section of road. This would be a lot of data. Very accurate.

There may be other options, but I can't think of any right now. What do you expect out of me at 7:00 AM in the morning anyways?!?!?

Modeler

That's too bad, because none of the options you've mentioned are actually implemented in the TL's Navigation system.