Accord 2003 - Strange Navigation Behavior

  • Thread starter Thread starter Ken
  • Start date Start date
K

Ken

Yesterday my navi was unable to correctly track my car. It was off by maybe
half a mile. After awhile it seemed to be ok.

Yesterday there was heavy cloud cover and storms. But I didn't think this
could affect GPS reception. Any ideas? (There was nothing covering the
antenna area in the rear.)
 
My 2000 tends to have reception problems when there is heavy overcast.
There is a weird icon in the upper corner of the screen. It changes color
to indicate the status of reception (red - none, amber - flaky, green -
good).
 
My 2000 tends to have reception problems when there is heavy overcast.
There is a weird icon in the upper corner of the screen. It changes color
to indicate the status of reception (red - none, amber - flaky, green -
good).
 
I have a hand-held GPS and it is much slower to acquire satelites when
it is wet and heavily overcast. I was in a hertz car with their
never-lost gps in fog and the unit couldn't pick up satelites at all.
(Interestingly, though my handheld worked.)

The accuracy of the GPS is related to how many satelites it is
reading. I was in boat the other day with clear weather and the unit
showed like 9 satelites. It seems to be able to get an initial fix
with 3 but sometimes the accuracy is low.

So, I wonder if maybe the unit was simply unable to acquire enough
satelites to get an accurate read.
 
I have a hand-held GPS and it is much slower to acquire satelites when
it is wet and heavily overcast. I was in a hertz car with their
never-lost gps in fog and the unit couldn't pick up satelites at all.
(Interestingly, though my handheld worked.)

The accuracy of the GPS is related to how many satelites it is
reading. I was in boat the other day with clear weather and the unit
showed like 9 satelites. It seems to be able to get an initial fix
with 3 but sometimes the accuracy is low.

So, I wonder if maybe the unit was simply unable to acquire enough
satelites to get an accurate read.
 
Moisture and especially heavy rain will effect satillite reception.
Probably true, but I'd like to share my experience driving through the
Brooklyn-Battery tunnel under the East River in New York a couple of
weeks ago. I was under a couple of hundred feet of water, a dozen
yards of dirt and rock, and a shell that was probably either solid
steel or five feet of rebar-inforced concrete. The Navigation System
tracked me every foot of the way.
Go figure.

Ron

Ron
 
Obviously, the GPS signal was lost when you entered the tunnel. This
is evidenced by the word GPS turning white from green. The navigation
system contains gyroscopic and speed sensors to keep track of the
direction of the car at all times. It was probably using dead
reckoning to follow the map for the time you were in the tunnel.

Dick
 
Very interesting. I've never noticed the color of the "GPS"
lettering, but I'll look for that. Where did you get the information
on this?

Obviously, the GPS signal was lost when you entered the tunnel. This
is evidenced by the word GPS turning white from green. The navigation
system contains gyroscopic and speed sensors to keep track of the
direction of the car at all times. It was probably using dead
reckoning to follow the map for the time you were in the tunnel.

Dick

Ron
 
I agree. The signal would have been lost. I'm curious, though about
the sensors you mention. My handheld simply extrapolates when it
experiences signal loss, ie, continues in the same direction at the
same speed. After some predetermined time, it goes into a holding
pattern and reports the lost signal. Obviously an auto-based system
*could* have a speed sensor, but I'm surprised if it does. Calculating
speed from the GPS readings should be much more accurate than the
speed sensor in the car and it provides a vector speed (direction and
ground speed).

I'm not saying you're wrong. I'm just curious about the source of your
information about the navigation system having a gyroscope and speed
sensor. I've never heard that before.

Tom
 
Most car-based navigation systems use gyroscopes and speed sensors in
addition the the GPS. ETAK (now Tele Atlas) built a working navigation
system back in the mid 80's without GPS. It used wheel sensors only
(gyroscope optional) and used the wheel sensors to calculate both speed
and direction of turn (direction found by different speed on left and
right wheel). It worked very well and even adjusted itself for tire wear
etc.

Even systems with GPS rely heavily on the speed sensor and gyroscope
since this data is available much sooner (the gyroscope will tell you
that you are turning WHILE you are turning. The GPS can only tell you
that you have turned after you have driven in the new direction long
enough to establish two nes GPS locations). The system will use GPS to
calibrate the speed sensor, so the speed sensor data is very accurate.

As a technicality, one could argue that the speed sensor really is a
distance sensor. It gives a number of pulses pr mile regardless of how
fast you are going. Unless you have some sort of time reference, the
"speed" sensors will only give you mileage.
 
This is all very fascinating. As a curiousity, what happens if different
size wheels are installed? Is there an adjustment to the settings possible?
As an example, the Accord coupe with 6 speed comes with 17 inch wheels
standard. The automatic has 16 inch wheels.
 
TL said:
I agree. The signal would have been lost. I'm curious, though about
the sensors you mention. My handheld simply extrapolates when it
experiences signal loss, ie, continues in the same direction at the

That's why the automotive units cost so much ;-)
speed from the GPS readings should be much more accurate than the

As long as they are available. The road sensors and turn indicators are
always "on".

Inertial navigation system in aircraft were pretty accurate long before GPS
came out.
 
That is quite impressive. Since starting this thread I noticed that the
letters "GPS" are sometimes white, which means no signal is being received.
This explains why the system still works fine with the gyroscopes and speed
sensors.
 
I agree that it has been a good discussion. I also was not aware of
the sensors in the system until the question was asked, and I started
looking into it. As far as cost, it seems justified when you look at
the equipment and capabilities. The main system is in the trunk
running a DVD with the data. The screen is a lot larger than the
handhelds too. Also, the voice input capability puts the car systems
in another league from the common GPS. The Honda system can take
about 135 voice input commands. Everything from changing radio
stations to setting the passenger-side temperature to entering the
phone number of a business you want a route to. The more I play with
it, the more I am impressed.

Dick
 

Latest threads

Back
Top