Shop OBEX P1 Docs P2 Docs Learn Events
GPS — Parallax Forums

GPS

geoldrgeoldr Posts: 15
edited 2008-11-23 20:04 in BASIC Stamp
Hello all, I am wondering about the parallax gps chip. How powerful is it? Say I have a robot that avoids collisions. All it does is roam around and avoid obstacles. With a GPS Chip, would I be able to tell it where to go? And while its going there it avoids anything that is in the way?

Comments

  • FranklinFranklin Posts: 4,747
    edited 2008-11-23 00:30
    Outside and within the limits of GPS (+/- 5m) yes it will

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    - Stephen
  • SRLMSRLM Posts: 5,045
    edited 2008-11-23 00:56
    Note that if you use GPS for obstacle avoidance, you have to have a really big playing field.
  • Carl HayesCarl Hayes Posts: 841
    edited 2008-11-23 01:08
    GPS will do exactly one thing:· tell you where on Earth the GPS antenna is.· If the antenna is on your device, it will tell you where your device is.· An inexpensive GPS receiver can usually do this within about 60 feet, sometimes better.· Expensive ones can be much more precise.· "Expensive" means tens of thousands of bucks.

    The use you make of the information is up to you.· You might, for example, use the "where it is" coordinates to calculate a direction and distance to "where you want to be", and if you've got a compass you can then steer to where you want to be.· But GPS doesn't do that; it only tells you where you are -- that is, where the GPS receiving antenna is.

    Oh -- GPS can do one other thing:· it can tell you what time it is, in principle within several nanoseconds.· The receiver has to know this in order to calculate the position, but most GPS receivers yield the time information up· to you very reluctantly, and very tardily.· Still, most GPS receivers will tell you the time with only a few seconds' error.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    · -- Carl, nn5i@arrl.net
  • gncguygncguy Posts: 35
    edited 2008-11-23 03:03
    This thread leads me to ask a question that has bugged me.· Since the GPS is that inaccurate why do people buy it?· A poor little boe bot could not keep up with 5 m errors.· I do believe that if you measure the position with the GPS sensor and then go back ask the robot to drive to that point the errors will be much less than the advertised GPS errors.· That is because most of the errors will be the same over short periods of time.·
    The use of GPS seems to imply the need for a high speed vehicle such as an rc car so that you can drive over football fields but as has been established in another thread there are no obstacle avoidance sensor good enough to support that speed.
    So what do people do with the GPS reciever.· I want to buy but I can't find anything to do with it other than build a GPS reciever.· Since those are so common now that does not seem that interesting.
    ·
  • SRLMSRLM Posts: 5,045
    edited 2008-11-23 03:15
    Like Carl Hayes said, you can use it to calculate a heading. If you are going to be going long distance (foot ball field maybe) then it'll do an admirable job of pointing you in the right direction. But think of it this way: your GPS is telling you where you are on a very big thing: the earth.

    The earth has a circumference(at the equator) of about 24,900 miles, or 131,472,000 feet. Your GPS can tell you where you are within 60 feet (the Parallax GPS seems to do it in about 10-20 in SoCal, USA). Even so, 60 out of 131472000 is pretty good. That's why GPS is best for large distances: because the scale of GPS is so large.

    Post Edited (SRLM) : 11/23/2008 6:09:21 AM GMT
  • Carl HayesCarl Hayes Posts: 841
    edited 2008-11-23 04:05
    Well, no -- Earth's diameter is a little less than 8000 miles.· Earth's circumference is about 24000 miles.· Still pretty big, though.

    Dunno about other people, but I bought my GPS to navigate in my airplane (back when I had one, can't afford that sort of thing any more).· Nowadays it keeps me from getting lost on trips·in my tiny little motor home.· Most of the time, anyway. turn.gif

    The GPS satellites transmit two codes.· The less accurate of these -- I forget what it's called, and I'm too lazy to look it up -- is the one nearly all civilian GPS units receive and use.· The other, more accurate, is called P-code, and it's encrypted.· Equipment that can decrypt and use P-code is rather closely controlled, but can achieve accuracy of the order of fractions of an inch.· I suppose if you had a P-code-capable GPS in your toys you could make them go from one corner of a room to the other without straying into the neighbor's house.

    Please forgive that word "toys" -- no offense -- I just can't bring myself to call anything a robot unless it's a lot like what Isaac Asimov wrote about, or at least what Karel čapek wrote about.· čapek, of course, invented the word.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    · -- Carl, nn5i@arrl.net
  • Carl HayesCarl Hayes Posts: 841
    edited 2008-11-23 04:10
    Incidentally, I am considering a purchase of the Parallax GPS for a project of my own.· I want to build a transmitter-finding system that will plot on a map the location of a hidden or rogue transmitter.

    It's easy to determine the direction from which a signal is coming.· Combine that with your current location, and you can plot a line-of-position for the rogue.· Do that from a car, and drive around a bit, and you can get multiple lines that cross.· Then you've got him.· Approach closer, and do it all again (or continuously), and your placement of his signal gets less and less uncertain.· I don't think a Stamp can handle the computational load, but perhaps a Propeller can.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    · -- Carl, nn5i@arrl.net
  • awesomeduckawesomeduck Posts: 87
    edited 2008-11-23 15:48
    The GPS module is a lot of fun. It was one of the first "upgrades" I bought. It only works outside, and it takes a minute to sync up, and its accurate in meters, not millimeters...but still its fun to play with. I don't think it makes much sense to use on a boebot, but I put the GPS and a serial LCD onto the PDB and had a working GPS unit in one afternoon. I put it in my car and drove around and found the speed readings to be accurate and heading to be pretty good to. The Long and Lat were also within 10 feet of where I was sitting...as compared to what Google Maps was saying. One thing I found was the altitude seemed low by 10 to 20%. The only reference point I had was from a survey of my house, which says my front door was at 402 ft...but I never read higher than about 380ft. Maybe my survey is wrong...who knows.

    I also found that I locked onto 4 to 6 sats pretty often, and the accuracy seemed to really improve the more sats I had locked.

    If you think its fascinating that you can take a couple of parts and stick them on a board, write a few lines of code and then be communicating with (or at least receiving from) a constellation of satellites 12,000 miles away then you should get the GPS module.
  • geoldrgeoldr Posts: 15
    edited 2008-11-23 18:12
    Hello everyone, and thank you for all the replies. Basically what I want to do with the GPS, is set waypoints and guide my robot car. I don't know what is best for this, GPS, compass, and whatever else. Say there is a track I want my car to drive on. I want to use the GPS to set waypoints on the track, so it completes it. Would the GPS module be capable of this?
  • Mike GreenMike Green Posts: 23,101
    edited 2008-11-23 18:30
    It all depends on the scale of the track and the car. Just keep in mind the accuracy of the readings. They're on the order of 20 to 50 feet. If your course is on the order of a mile or miles, a GPS receiver will be very useful for marking waypoints. If the course is 50 to 100 feet in size, a GPS receiver will be of little use. It'll tell you which end of the course your car is on, but that's about all.
  • hank van burenhank van buren Posts: 57
    edited 2008-11-23 19:26
    I have been able to hold +/- 2 feet Using a Garmin OEM 15H, WAAS enabled www.gpscontrolledvehicle.net/ Hank
  • gncguygncguy Posts: 35
    edited 2008-11-23 20:04
    What has been said about GPS accuracy is too except for one small subtlety. If you drove over a course and have the computer record all of the waypoints. You could then let the car go and it will drive over the exact same point within inches. this is possible because the GPS points were surveyed ( measured by the GPS unit) within minutes of when the car drove the course. The GPS errors are biases that vary at different frequencies. the high frequenices errors seem to be on the order of inches. Since the car is doing relative navigation to its own measurements only these short term variations show up in the rapid run. If you came back a week later, all of the biases would be on the order of the often quoted GPS errors for the particular reciever. the first case was relative navigation and the second is absolute navigation. the first case would only be of interest for toy or hobby purposes.
Sign In or Register to comment.