Page 3 of 3   << first < 1 2 3
Bookmark this page: Add Just now smart are these things to Yahoo MyWeb Add Just now smart are these things to Google Bookmarks Add Just now smart are these things to Windows Live Add Just now smart are these things to Del.icio.us Digg Just now smart are these things ! Add Just now smart are these things to Netscape
  •  
  • Subject
  • Author
  • Date
If you were  Registered and logged in, you could reply and use other advanced thread options
Posted by Howard Lester on December 2, 2009, 6:42 am


"Howard Lester" wrote

> "Fred McKenzie" wrote
>> As TulsaOK suggested, you might take the trip again and see if the
>> alternate route had indeed been "learned".

> I did that with my Nuvi 350, and the answer for it is "no." In fact/in
> addition, the Nuvi acknowledges that my alternate route shaves one minute
> off the arrival time. I see that on the screen immediately after Jill lets
> me know she's recalculating. I thought, after the first time I took the
> alternate, that I might have had the Nuvi set for "shortest distance," but
> no, it was set for "shortest time." So trying it again I got the same
> result.

To follow up: apparently the third time's the charm. The Nuvi directed me
via the "new" route.



Posted by who where on November 22, 2009, 8:19 pm


wrote:

>> Still in familiar territory, it wanted to send me around two sides of
>> a triangle. I chose the third side, which was shown on its map and
>> was also shorter than either of the others. On the return journey
>> "destination: home" it used what it had learned and sent me back the
>> road I had used.
>>
>> Make what you will of that, and be sceptical if you wish, but it did
>> "learn" a shorter route.
>Who-
>As TulsaOK suggested, you might take the trip again and see if the
>alternate route had indeed been "learned".

I will certainly watch what it does as and when I visit him again.
Alternatively, I could try the return journey first and - assuming it
uses the short path - then do the outbound.

>Somewhere I read that the algorithms were designed to prefer right turns
>and avoid left turns. Could that have made a difference in your
>triangle?

Transposing left for right in the above (we are in Oz, driving on the
left ...) yes it would have - the short path was a right turn.

>I have experienced a bias against making U-Turns. In one case a
>destination was on the left. It would have been necessary to go beyond
>it and make a U-Turn. I was routed to the right, then to the left three
>times to arrive at my destination on the right.

I haven't found a U-turn bias. What I have found though is a liking
for calling for a U-turn if you overshoot a destination on a
junctionless road. And sometimes the nominated U-turn location beyond
the target is uncanny - a gateway into a large empty paddock. How the
heck do they get THAT level of detail into the database?

Posted by Mike Coon on November 23, 2009, 5:01 am


who where wrote:
> ... And sometimes the nominated U-turn location beyond
> the target is uncanny - a gateway into a large empty paddock. How the
> heck do they get THAT level of detail into the database?

Alternatively someone could have built the paddock to assist all the people
who get told by their Satnavs to do a U-turn just there...

Mike.
--
If reply address is invalid, remove spurious "@" and substitute "plus"
where needed.



Posted by Mike Lane on November 23, 2009, 8:31 am


who where wrote on Nov 23, 2009:

> I haven't found a U-turn bias. What I have found though is a liking
> for calling for a U-turn if you overshoot a destination on a
> junctionless road. And sometimes the nominated U-turn location beyond
> the target is uncanny - a gateway into a large empty paddock. How the
> heck do they get THAT level of detail into the database?

They collect an astonishing amount of data for each road that is surveyed -
up to 214 separate attributes are collected, I understand.

I take it you've seen the Navteq surveying film?
http://tinyurl.com/3ow8cl


--
Mike Lane
UK North Yorkshire
email: mike_lane at mac dot com


Posted by dold on November 23, 2009, 1:27 am


> On the return journey
> "destination: home" it used what it had learned and sent me back the
> road I had used.

It just had a different idea of the proper route for two directions on what
you think of as being the same road.

Between Lower Lake and Middletown, in California, a Northbound route looks
correct, but a Southbound route will divert to a side road for a couple of
miles and come back to the highway near Hidden Valley Lake. I presume that
the side road was the highway once upon a time.

I've driven the road many times, and the GPS hasn't learned anything about
the faster route.

--
Clarence A Dold - Hidden Valley Lake, CA, USA GPS: 38.8,-122.5

Page 3 of 3   << first < 1 2 3
Related Posts
  • Subject
  • Posted In
  • Date