GPS and Compass

[cmartin]

I have several questions about how to use GPS and a digital compass for robotics.

**broken link removed**
I have a Trimble, Lassen SK II GPS module. I am considering buying this antenna for it **broken link removed**. The antenna is 5m long, which is more than I need. What do I do with the excess cabling? I don't have a connector and crimper to cut the cable down. Is it OK to coil up the excess cable? How should the the antenna be mounted? Does it have to be on a pole above the robot?

I know I need a compass to tell me what direction I am going, GPS only tells you where you are. I don't know what compass to buy. I found this one SparkFun Electronics - Compass Module - HMC6352, is it good enough for basic usage? Will the magnet from the GPS antenna mount, motors, or power regulators interfere withe the compass? The robot will be in mixed terrain: grassy incline, sidewalk, flat grass field, dirt and bark, but no extreme conditions. Will bumping up and down throw off the compass readings drastically? SparkFun Electronics - Compass Module with Tilt Compensation - HMC6343 would take care of that problem but its $100 more than I would want to spend. Also where do I put or mount the compass? I have room inside the case to put a 1"x1" board or does it need to be mounted out side the case?.
**broken link removed**

Thanks for the help.

 

[Diver300]

I have several questions about how to use GPS and a digital compass for robotics.

**broken link removed**
I have a Trimble, Lassen SK II GPS module. I am considering buying this antenna for it **broken link removed**. The antenna is 5m long, which is more than I need. What do I do with the excess cabling? I don't have a connector and crimper to cut the cable down. Is it OK to coil up the excess cable? How should the the antenna be mounted? Does it have to be on a pole above the robot?

That antenna looks like it should work. Just about all GPS antennas are amplified and are fed from 3V or so from the receiver.

The 5m of cable is bad for sensitivity.

RG174 Coaxial Cable | Wellshow Technology shows about 2 dB per meter loss, so 90% of you signal will be gone by the time you get to the receiver. I suggest that you cut and join the cable. Even if your joint is poor, it won't loose 90% of the signal. Otherwise, just coil it up.

You should mount the receiver on the top of the robot. There is little advantage of having it on a pole, unless you are somewhere where a couple of meters up would give you much better signal, like in a cornfield.

 

[Oznog]

I'm not so sure. The antenna is surely amplified- which generally negates any problem with loss.

However, that level of amplification may have been adjusted to work with that length of cable. If the cable was reducing the signal by 10db, then the antenna may be providing 10db of gain to create an optimum signal at the GPS. If the cable were removed, and the antenna gain not lowered, then the receiver will see a signal 10db higher than designed, which might be able to overload the receiver and cause bad reception.

 

[Vizier87]

Talking about HMC6352, are there replacements for the cheaper KMZ51, since it's discontinued from production?

 

[cmartin]

The below is from the data sheet for the GPS module in the Accessories section.

Compact Magnetic-Mount 5V GPS Antenna
Compact 5V magnetic-mount, active micropatch antenna with 5 meter cable. SMA or SMB connector. 27 dB or 35 dB gain.

And the picture that goes with the description looks like the picture provided in the ebay auction. If the data sheet list that antenna then it should be fine to use. I also found a matching antenna that has a 1m long cable and will try that one too.


From the HMC6352 data sheet features section.

Sensor Can Be Used in Strong Magnetic Field Environments
Stray Magnetic Field Protection and Temperature Compensation

And other text leads me to believe that I should not worry about the compass being effected by other parts in the robot.


From the HMC6352 data sheet.

At worst case, each degree of tilt from a level orientation could
add two degrees of compass heading error.

The play field for the robot has mixed incline/declines of more than one degree. Tilt error will be a problem. To compensate, could a three axis accelerometer or a X, Z gyro be used to determine tilt while moving? Also, could the either one aide in movement control?