Autorouting questions

[K2R]

I would like someone to shed some light on autorouting and verify what I have found as this is new to me.

As motorcyclists, we often ride routes that crisscross. For example, lets assume one builds a figure 8 route on MapSource and downloads it to a 2610. Assume the start and end points are at your home.

I believe the 2610 will follow the exact route as downloaded so long as you don't deviate from it, i.e. don't get "off-route". Once off-route things change as follows:

If the Off-Route Recalculation option is set to Automatic, the 2610 will re-calculate the route, and direct you home (by shortest/best etc. which will again depend on settings) and completely ignore your planned route.

If the Off-Route Recalculation option is set to Off, the original route is maintained but you will still be directed to follow the best/shortest path home.

To illustrate, assume you get off-route just before the first cross over point of the figure 8 route. Either option will direct you to ignore the top part of the 8.

Am I correct? If so I would sure like to have an option to follow the specified route regardless.

 

[Timolgra]

Now I'm a GPS dunderhead but surely you mark the cross over point as a 'Waypoint' and/or you go 'Via' certain 'Waypoints'.

There told you I was a dunderhead.

'Tim'

 

[BurnieM]

lets assume one builds a figure 8 route on MapSource and downloads it to a 2610. Assume the start and end points are at your home.

Is this correct ?

I thought that you could not load a manual route to a SP2610 (or SPIII).

I believe the behaviour is that the unit will re-calculate the route to the destination as soon as you tell it to go there.

 

[roadrider]

As motorcyclists, we often ride routes that crisscross. For example, lets assume one builds a figure 8 route on MapSource and downloads it to a 2610. Assume the start and end points are at your home

If you go off route, the gps will do up to 3 quick recalculations to get you back onto your route, (thanks Pan Euro) if it fails to get you back on route after the 3 attempts, it will do a full recalculation of the route. is this what your asking??
John

 

[K2R]

roadrider, you are referring to this "The GPSR now does 3 "quick" recalculations - recalculations that only consider how to get you from your present position to the closest point on your pre-existing route - and if you don't get re-established on your pre-exisiting route after the third "quick" recalculation, the GPSR gives up on you, and does a "complete" recalculation, from your present postion all the way to your destination."

His explanation would seem to support the 2610 trying to keep on route. I suppose I'm talking about the case where you have a nice long ride laid out that crisscrosses itself. You get off-route to gas up. You start back. The GPS looks at where you are, where you want to eventually end up (at home) and says great, does a recalculate and as a result cuts out most all of your planned downloaded route in the belief you want to go home now. (Note that the route leg going back home is part of the pre-existing route.)

I quite sure that's what it did to me on a ride yesterday but I could be mistaken. I thought I could prevent this from happening by turning off auto-recalculation. But I don't think that solves my problem. I'm confused obviously.

 

[PanEuropean]

Interesting questions. I'll try to give you an answer that explains what is happening (kind of long-ish answer) rather than just an answer that tells you how to fix the problem.

First of all: Burnie:, sure, you can create your own routes and download them to the SP 26xx series, same as you can for any other Garmin GPSR. All you have to watch out for (same as on the SP III) is that you have the same cartographic product visible on your PC screen as you have loaded in your GPSR. Otherwise, you'll get a warning message on your GPSR when you try to use the route, telling you it was created using maps that are not loaded on the GPSR.

What I mean by this is: Let's say you have both CN and CS Europe loaded on your PC - or both CN 5 and CN 6 loaded on your PC. Check the little box at the top of the MapSource screen on your PC to make sure that the map you have visible (active) is the same as the map you have loaded into your GPSR. Otherwise, all your work will be for naught, the GPSR will recalculate your route when you first attempt to make it active.

Second: Make sure that the preferences you have set for routing on the PC - from the MapSource preferences control (EDIT menu, then PREFERENCES at the bottom, then ROUTING tab) match the preferences you have set in your SP 26xx (MENU button, then OPTIONS, then ROUTING). Doing this will save quite a bit of work down the road, and give you much more consistent results.

Third: Understand that the GPSR will, all other things being equal, try to calculate the most efficient route between A and B. This is only common sense, it's why people buy auto-routing GPSR's. If the route you are trying to calculate is not, by definition, a sensible or efficient one (and a figure 8 route does not meet either of these criteria), then you will have to take a few extra measures to get the results you want.

There are several ways to address the 'figure 8' route problem. First, and probably simplest, is to break the route into several segments, so you don't have route overlaps. By example, if the start and end points are at top dead centre of the 8, make one route that leads from top dead centre to bottom dead center, and a second route that leads back up again, from BDC to TDC. This eliminates the overlap problem. What you now have is two 'S' shaped routes (one will be like a backwards-facing S). When you get to the end of the first route, just activate the second route.

The second way - more difficult to do, more time consuming, and fraught with more possibilities for error - is to insert enough VIA points into the route that you pretty much eliminate the possibility of the GPSR choosing (when recalculating) any path other than the path you want to follow. If you then get off-route, the GPSR will do 'quick' recalculations to get you back onto the active leg of the route that leads you to the next VIA. If you have not got back onto the active leg by the time the GPSR has done 3 quick recalculations, it will do a full recalculation that will create a new route from present position to the currently active VIA. The only exception to this will be if you have already passed abeam the active VIA, in which case it will recalculate a new route to take you to the next VIA in the sequence.

For further information about the route recalculation process that K2R refers to in his post above, see my original post about recalculations at this link: Bigger Data Card (Feb 26, 2004 post - PanEuropean)

Regards,

PanEuropean

 

[K2R]

Pan European. I'm going to do some more experimenting. I understand the concept of what the GPS wants to do, basically to take you to your end destination, and from a routing perspective, is not interested in where you have been. I believe that if you turn auto-recalc off, and then proceed along getting back on your chosen route, it will pick up and direct you where you want to go. Realize the figure 8 is just about the simplest example of this situation. I ride many roads in unfamilier territory that cross one another and hate to take the time to figure out which way to go at every crossing. As I said, the motorcyclist often operates under different parameters than other drivers.

I also have MetroGuide Europe where I have built a number of routes. Not being an autorouting product, I would not expect to encounter the problem described.

 

[PanEuropean]

"I believe that if you turn auto-recalc off, and then proceed along getting back on your chosen route, it will pick up and direct you where you want to go."

Yes, this is correct - although if you have auto-recalculation off, you do give up one fairly important capability of the GPSR, which is to get you back on your desired track with a minimum of effort, should you stray off it either by choice (traffic congestion, unsuitable road surface, etc.) or you are forced to stray off it (error in the cartographic database - road not there).

The trick, when laying out routes that are not de facto the most logical route from A to B, is to insert sufficient VIA points (or just insert sufficient plain old WAYPOINTS) that the GPSR, when recalculating, will decide that the route you have chosen is, in fact, the most logical route between the WAYPOINTS or the VIA points.

I also ride a motorcycle, and now have 3 years and 80,000+ km's experience using both the SP III and SP 2650 for routing. Here in Europe, where I do most of my riding, I often lay out routes that are circular - for example, from my base near Zürich out through the nearbly Alpine passes and then back home. Other than some initial problems when I was first getting used to the software and hardware (which I think is what you are going through now), I have never had any difficulty getting the GPSR to take me where I want it to - with auto-recalculation turned on.

 

[BurnieM]

Okay so you need multiple via points to follow a non-direct autoroute.

Am I correct in thinking that the GPSR recalculates the route itself (using any via points you put in) and does not just take the route calculated on the PC.

Obviously, if the same mapping product is used on both then the route should be the same.

 

[(RIP) Dennis]

Okay so you need multiple via points to follow a non-direct autoroute.

If I may add my two pennyworth... I would also suggest that you enter the route as shorter legs, rather than as one long contiuous route. This gives you a couple of important advantages.

1. Any recalculation done is much quicker, just a few seconds, whereas on a long route you may well havew passed a suitable turn-off before the recaculation has finished

2. You always know where you are going to end up. One slight problem of GPS is that you are blindly following a dumb computer assuming that it knows what it is doing, and just as importantly, assuming that you have entered the correct imformation.

It only takes a couple of seconds to pick up the next route.

 

[PanEuropean]

Burnie wrote:

"Am I correct in thinking that the GPSR recalculates the route itself (using any via points you put in) and does not just take the route calculated on the PC."

Yes, Burnie, you are correct, this is exactly what happens. When you construct a route on the PC, using the MapSource application, what is saved is the following:

1) Origin and destination

2) Visible WAYPOINTS (e.g. waypoints with a name) and 'less visible' VIA points (e.g. intermediate points you have created by dragging the route line around with the pointer tool and mouse).

2a) Totally invisible turn points that the MapSource application creates to demarcate certain turns.

3) Name and version number of the cartographic product used to create the route.

Note that routing preferences stored on the PC are not saved.

When you transfer this route to your GPSR and then activate it, the GPSR recalculates the route using items 1, 2, 2a and 3. If the routing preferences are the same on the GPSR, then the result (route displayed) should be identical. If the routing preferences are different, the route may vary somewhat between the VAYPOINTS or the VIA points. If the routing preferences are the same, but the route displayed on the GPSR is different than the route on the PC, you need to increase the granularity of the waypoint or via point placement, to constrain the GPSR to the path you want to follow.

Dennis, you are 100% correct with your comment about shorter routes. Recalculation time is less of a concern with the newer units (and newer software), but in principle, the shorter the route, the fewer are the possible problems you could encounter in reproducing it on the GPSR.

Again - if your route is simple, for example, New York to Los Angeles, then you can do it in one route segment. But if your route is complex, for example, home to the milk store via the airport, the bar, the library, and the moto dealer, then it's best to break it into a few chunks, especially if the route does not (overal) keep going in the same direction - meaning, if there are any course reversals or double-back points, then break it into two where those happen.

 

[Timolgra]

Yes, but still a dunderhead.

Tim

 

[K2R]

my intent

I appreciate the feedback on this thread. Just a mention that my intent with a thread such as this is only to gain a better understanding of the capabilities of the device so that I can use it most effectively. Basically I can't find my way around the block w/o getting lost.

I tried a few routes today. Without going into detail, it does make the attempt take you to missed vias. This can get a little convoluted but it verifies what Pan Euro said.

I have been using GPS since 1999 but the 2610 is my first experience with autorouting. The 2610 is quite an amazing device. Its nice to have a separate discussion topic focused on GPS.

 

[roadrider]

Re: my intent

I appreciate the feedback on this thread. Just a mention that my intent with a thread such as this is only to gain a better understanding of the capabilities of the device so that I can use it most effectively. [/QUOTE
KR2,
There's a lot of very good discussion on the GPS threads. And I,m certainly understanding mine much more than I did last year.
John

 

[PanEuropean]

A funny story here, just to let all of you know that these "routing problems" are not limited to Garmin GPSR's.

I finished training and examining a crew this evening in a simulator, and as we were going out the door, the crew asked if they could "have a spin" in the MD-11 simulator in the next building. This simulator will be taken out of service in a few months, when the MD-11's in the fleet are replaced by A-330 and A-340 aircraft, so I said 'sure' and we all climbed into the MD-11.

The flight management system in the MD-11 is not quite the same as the FMS I am used to, but it is similar. However, I'm not an expert at it. All I wanted to do was program a takeoff from Geneva runway 23, a short hop out to the initial approach fix of the runway we departed from (SPR VOR), then an VOR-DME approach back in to land on the same runway.

Once we were airborne and turning downwind to the VOR (about 15 miles away), I tried to set up the approach. But, we were heading north-east to the initial approach fix, and the approach itself runs south-west (makes sense, because it would take us back to where we originated). This is almost exactly the same "figure 8 routing problem" that Bill (K2R) referred to in his post at the top of this thread. There's no published procedure turn for this approach, all the Swiss approaches these days are "straight-ins" and you use a STAR (standard terminal arrival route) if you are coming from the wrong direction. I didn't want to get into setting up a STAR prior to the approach, otherwise, it would have taken us half an hour to fly the darn thing, and I wanted to get back to the hotel before the kitchen closed.

Do you think I could get the aircraft to accept this takeoff, downwind turn, and 'direct the VOR, straight in to 23'? Like hell I could. Every time I entered the approach and activated it, the aircraft wanted to "pull a U-turn" and head back to the airport. Finally, I just said to heck with it, and waited till we were long past the IAF, going towards Zürich - at that point, the crew turned the aircraft around by hand, we activated the approach, and all went well.

I am sure I could have done what I wanted to do if I was an expert with the FMS in the MD-11 - but this experience just goes to prove that if you want to ask a navigational system (any navigational system, be it a MD-11 FMS or a motorcycle GPS) to do something unusual, you need to know exactly how the system works, and be very precise in your request.

PanEuropean