Matrix user forums

Hello all,

I need to use a GPS for tracking a very low speed vehicle (1-100 meters/hour).

I actually need to measure the distance between two points which one of them is a fixed point and the second is a moving point.

I also need to monitor the current speed of the moving point at all times.

I need to gain the best accuracy that i can which is below 1 meter if possible.

I am wondering if there is a chance to achieve all the above and if there is, what kind of GPS system should i use for this purpose?

The microcontrollers that i can deal with are all from microchip including DSP devices for this project.


Waiting other idear or suggestions.

Thanks in advance
George

Hi George,

Sounds like you need a GPS module with a large number of channels, from memory some of the Telit modules have up to 20 active channels but this could be wrong.

The more channels you have available, the more satellites you can see and hence the more accurate your positioning, ground speed is measured as part of the positioning algorithm and available in the streamed data from the GPS module.

Note that some areas / governments actually obfuscate or alter GPS signals to ensure that accuracy is poor or offset so it matters where the system will be running.

It may be worth getting an GPS position and then using some form of secondary feedback to help aid the positioning. E.g. accurate encoders on the vehicle wheels to tell how far the wheels have travelled. Regular IR beacons along the route etc...

Hi Benj,

Thanks for your reply!

unfortunately it is not possible to use encoder for measuring the distance at this time, that is why i need something like GPS or so. If i was able to use encoders i could easily measure the speed of the vehicle as well.

I tried to see if there is any other way of getting the distance from a fixed point and a moving point without GPS but i have no results .

Note that between the two points there is not visual contact.

Do you thing that it would be possible to measure distance using RF, ultrasonic or anything like this ?

thank you very much
George

Could you use markings on the ground maybe, or does the route change etc?

The route changes.. unfortunately neither this is possible.

Some sort of ultrasonic ping might work, if you can make it loud enough to travel the distance.

You would need to have a way of synchronising the moving point with the non moving point so that the distance can be measured e.g. a wire between the two.

e.g. the electric wire goes from high to low. The moving point listens for the falling edge and sends out an ultrasonic ping. The fixed point listens for the falling edge and starts timing until the ping is received.

Using the time difference from the falling edge to the ping based on the speed of sound you can work out the distance.

Your timing resolution will determine the accuracy of the returned value.

If the distance has to be really far then you may have to start taking the speed of light and transmission principals into the equation to work out the time difference along the cable. Fibre optics would help get rid of the transmission principals such as inductance and capacitance effects on the line.

Benj wrote:Some sort of ultrasonic ping might work, if you can make it loud enough to travel the distance.

You would need to have a way of synchronising the moving point with the non moving point so that the distance can be measured e.g. a wire between the two.

e.g. the electric wire goes from high to low. The moving point listens for the falling edge and sends out an ultrasonic ping. The fixed point listens for the falling edge and starts timing until the ping is received.

Using the time difference from the falling edge to the ping based on the speed of sound you can work out the distance.

Your timing resolution will determine the accuracy of the returned value.

If the distance has to be really far then you may have to start taking the speed of light and transmission principals into the equation to work out the time difference along the cable. Fibre optics would help get rid of the transmission principals such as inductance and capacitance effects on the line.

Thank you Benj, i really appreciate your help

No bad idea to use ultrasound transition!

According your scenario, there is one problem regarding the wire or the cable. It is not possible to have the vehicle(moving point) wired and move it 600m away of the non moving point. To overcome this, i can modify a little bit your scenario and instead of the wire, to have a double deep sound separated by a fixed dead delay.

BEEP___________BEEP__________________________________________BEEP___________BEEP__________________________________________...


Using this technique i will be able to start measuring the time from the moment that the first BEEP arrives until the second one and i will do the calculations..



I will work this more in my mind and i will come up again with more details and ideas. Any suggestions will be appreciated.

Any idea of what kind of ultrasonic transceiver i could use for this??

Thanks in advance
George