GPS, car direction

[Balthisar]

@scr4

If you had multiple antennas, it may be easier to use the phase difference to calculate which direction the radio signals are coming from.

You don’t need anything as fancy as measuring phase differences. Just calculate the position of each antenna, and geometrically derive the heading. In the end, you’re not calculating the position of the GPS device anyway, but its antenna(s).

[CJJ]

@Xema

I did some googling and there are indeed units that contain “gyros” for dead-reckoning navigation. I put the word in quotes because in all the examples I found these were solid-state inertial sensors (piezoelectric or fiber-optic);

I have personally worked on the HW and SW for a major auto company’s fleet of GPS navigaton systems. Dead reckoning is used to assist GPS navigation on most of their vehicles (urban areas in particular rely on it because GPS reception can be spotty in some areas and the distance between streets is rather short). This involves data sent from wheel sensors to the GPS, usually over the vehicle data bus.

One other point: If the vehicle is off for a long-enough period of time, relying on GPS alone for position and heading would mean waiting a minute or so for navigation to be available (you need at least 3 satellites for a 2D fix). Again, last-known position and whell-sensor dead reckoning provide the driver with nearly instant access to navigation.

[1010011010]

@scr4

If you had multiple antennas, it may be easier to use the phase difference to calculate which direction the radio signals are coming from.
Actually I’ve always wondered how difficult it would be to implement this method. Does anyone know if it’s been done?

Large merchants that autonavigate using GPS waypoints will have receivers at the bow and stern for GPS derived heading. This is based off the calculated positions of the front and back of the boat, not “phase difference”.

Also, the GPS satellites broadcast the time, their own positions, as well future orbital data as part of their individual signals. Each satellite in the constellation has a different signal overlay so they can be identified and differentiated.

[Kevbo]

@Rick

A single GPS antenna. (Sorry Kevbo just one not two)

No need to be sorry. I was not clear in my post. I was only speaking to the claims that it is impossible to get heading information using only GPS. It is possible, but that is not how it is done.

[dynamitedave]

@scr4

If you had multiple antennas, it may be easier to use the phase difference to calculate which direction the radio signals are coming from.

Actually I’ve always wondered how difficult it would be to implement this method. Does anyone know if it’s been done?

Tractors with basic auto steering have one antenna. If the machine is being used on hills and the accuracy required demands slope correction, to correct the error one system is to use 3 antennas in a triangle, one at the front centre of the cab, the other two at the rear corners. The software evaluates the height difference between the antennas and applies an offset to the programed course.

:smack: :smack: nevermind, totally irrelevant to the the above query and to the thread in total as I don’t know if the system is capable of supplying heading info.

[suranyi]

@SmackFu

I’ve noticed my TomTom cheats. It assumes I’m following the route it suggests, and just locks me to the road. If I take an exit it isn’t expecting, it will show me on the highway for a bit, pointing in the road direction for a few seconds, until it notices I’m too far off the route and moves me. It’s like I’m on rails.

This is typcial behavior of a map-matching algorithm.

(I also work for a company that creates digital road maps for navigation systems.)

Ed