The Iris will not arm itself with DOP above 2.0, even with 9 sats and "3D GPS" displayed on telemetry. The screen reports high H (horizontal) DOP, possibly because I am surrounded by hills. Can the DOP threshold be increased?
You need to be a member of diydrones to add comments!
Bill,
Thanks for that explanation, however I just went out to test that and could not get an HDop below 2 and the IRIS would not allow me to take off in stab mode. I double checked several times and the GCS continuously announced preflight failure because of high HDop. I raised the acceptable HDop in the advanced params and it turned green, armed and allowed take off.
Something's not quite right. My iris is set to default params, other than the above param.
Thanks Bill, that was a very nice explanation and makes a bit more sense now. The HDOP was throwing me off and confusing. I just came back from my first real flight with the Iris. It went perfect. I thought there was a problem as it was not going where I wanted, but I had switch 7 on and it was moving according to my position. But once that was off I kept an eye on the GPS HDOP and it was very low around 140 I have to now start to understand more about Mission Planner so I can take advantage of things. Is this my best choice or would I be better of with one of the others. And if I did change, it will not effect my saved setting in the pixhawk right?
It was so nice watching it fly and it handled very nice. I just got mixed up some on the Alt vs loiter but it all worked out and I was ready to push up to STB should something go wrong. Wow this sure takes some worries away. You are ok Bill
A few of us are having this issue I am awaiting a GPS shipment so I can swapout a supposedly defective GPS...(it gets hdop of 3.7 and higher rarely below 2.1) in a clear sky while my iphone 4S and android GCS are reporting 8 satellites in a basement room with 1 floor and a roof above that overhead with a 3-5 meter fix..
WTF!
Same symptoms in a wide open park no trees or sky cover(berkeley marina where 3DR employees fly!!).. a bunch of experienced non 3DR drone pilots were being very watchful when I attempted to switch to GPS modes(because of the high HDOP IRIS was not very predictable.)
After GPS swapout if this continues I am digging into the code that implements the HDOP math and checking it for errors(I do kind of know this stuff).
As I suspect we may be seeing a bug in the APM firmware built for the iris, I will also use a utility to chat with the ublox directly and see if its telling me a different story than the pixhawk but NOT doing that until I see the swapout do the same thing , at this point I think we may be looking at a few flaky ublox chips(NOT unheard of from other vendors also).
HZL
ps you are very correct in being extremely cautious with the IRIS until this is resolved as it it so heavily dependent on a properly working GPS.
I don't think it is just an IRIS problem. I fly an APM board with a different GPS unit and have experienced the same issue. A flying buddy of mine flies NAZA and has the same issue. When I have an issue he has the issue. It will come down, but sometimes it takes 10 minutes or so. I'm guessing it has to do with what satellites are visible at the time, and how they are being moved around right now to accommodate some out of service ones, and also the new ones being launched. If a bunch of the satellites visible to you are near the horizon, the antenna on the ground will have a hard time picking them up. I also have notice when I am flying my full scale plane with the latest GPS software and hardware I have been getting degraded signal messages pretty frequently. If I go a few minutes and the HDOP is stuck and not dropping I unplug and restart the APM. Seems to change something and get it to lower the HDOP quicker. Also I have noticed once it has found itself subsequent restarts happen almost immediately. I am pretty certain its a GPS issue with the sats and not the actual IRIS. When I have witnessed 3 completely different systems in the same day, around the same times having the same issue using different antennas, different firmware, and different software. I started a thread about it more than a month ago but it got buried with lack of responses.
hmm but that doesnt explain why both iphone 4S and Android(GTP313 were getting 3 and 4 meter fixes in a basement with wifi turned off(I have loaded up utilities that allow me to see the satellite constellation numbers of satellites and precision of HDOP and VDOP and other navigation fixes)
while Ublox 6H in the IRIS with or without a copper ground plane is seeing same exact satellites BUT cant get HDOP below 2.1 rarely... if NOT defective GPS chip antenna then we are looking at some serious desensitization here or software errors in how HDOP is calculated.. hmm if HDOP is same issue on NAZA that would point to same code with same bug, does NAZA use the Ublox 6H by per chance? I am tracking accuracy of HDOP on 3 separate devices and logging and the IRIS using the ublox 6H seems to be having the issue.. not the Satellite constellation, so next after GPS swap for me is a utility to directly log the fixes coming from the Ublox 6H and log those along side the iphone 4S GPS sensor log and the same for the android gtp3113. Have to see if HDOP calculated in the APM code or in the UBLOX 6H directly(I suspect the latter).
Until my navspark chips arrive later this month(hopefully) I dont have a comparison independent GPS chip to test these suppositions against..(at this rate I will hook up a navspark in GLONASS mode for a sanity fix on these GPS issues.)
HZL
ps ordering 2 RAW mode navspark for GCS usage so can have accurate GCS fix and xmit corrections with software I have yet to develop
hotelzululima > hotelzululimaMarch 20, 2014 at 8:07pm
seems to get a LOT more accurate fixes than I do from the exact same GPS but on a bread board..
This discussion here seems to touts the NEO 6M as getting 1.3 HDOP under a metal roof even though a cheaper GPS(one of those arriving monday from goodluckbuy)
its work a try to get things working.. even a much cheaper unit...
HZL
Evaluating the 3DR uBlox LEA-6H GPS - Wayne's Tinkering Page
I like to design and build various kinds of electronic gadgets and this site is my way of sharing my work so that others can benefit and learn from m…
hotelzululima > hotelzululimaMarch 20, 2014 at 8:40pm
the second discussion with http://diydrones.com/forum/topic/listForContributor?user=3dsw9jz7wfl9y seem to point to a another solution.. slow down the update rates to 2-4 hz get 10-14 sats instead.. set mode for pedestrian on the 6H to improve loiter and prevent GPS wander he has a config file and APM code changes in that discussion.. I will be looking into the code he is pointing at in APM.
HZL
ps the MTK GPS is mentioned also as getting 10-14 sats and not getting bad HDOP
I get 1.3 routinely once it captures. Cell phones also cheat because they use the towers to get a quick position so they know the area they are in before they start looking for sats. I don't know much about the codes or refresh rate issues. I am not even close to a programmer and don't know squat about code. But once captured my gps works great.
actually the utility I m using is accessing the GPS device directly so no the cell towers DONT read into the utilities display of HDOP(which is coming from the GPS only.. cell tower access has no HDOP equivalent) and cant on the android device as it is only wifi and that turned off. My GPS is defective from what I can tell. am looking for a PIXHAWK passthru utility so I can use ucenter to chat with the GPS directly and need my new Micro USB hull port as that broke first thing to be delivered to do any of this(replacement GPS in same package).
Regarding the video about erratic throttle behavior when motors are started in loiter mode, I have had a few surprise behaviors in that mode but it never occurs when armed in stabilize mode.
Replies
Thanks for that explanation, however I just went out to test that and could not get an HDop below 2 and the IRIS would not allow me to take off in stab mode. I double checked several times and the GCS continuously announced preflight failure because of high HDop. I raised the acceptable HDop in the advanced params and it turned green, armed and allowed take off.
Something's not quite right. My iris is set to default params, other than the above param.
Thanks Bill, that was a very nice explanation and makes a bit more sense now. The HDOP was throwing me off and confusing. I just came back from my first real flight with the Iris. It went perfect. I thought there was a problem as it was not going where I wanted, but I had switch 7 on and it was moving according to my position. But once that was off I kept an eye on the GPS HDOP and it was very low around 140 I have to now start to understand more about Mission Planner so I can take advantage of things. Is this my best choice or would I be better of with one of the others. And if I did change, it will not effect my saved setting in the pixhawk right?
It was so nice watching it fly and it handled very nice. I just got mixed up some on the Alt vs loiter but it all worked out and I was ready to push up to STB should something go wrong. Wow this sure takes some worries away. You are ok Bill
Take care
Michael
A few of us are having this issue I am awaiting a GPS shipment so I can swapout a supposedly defective GPS...(it gets hdop of 3.7 and higher rarely below 2.1) in a clear sky while my iphone 4S and android GCS are reporting 8 satellites in a basement room with 1 floor and a roof above that overhead with a 3-5 meter fix..
WTF!
Same symptoms in a wide open park no trees or sky cover(berkeley marina where 3DR employees fly!!).. a bunch of experienced non 3DR drone pilots were being very watchful when I attempted to switch to GPS modes(because of the high HDOP IRIS was not very predictable.)
After GPS swapout if this continues I am digging into the code that implements the HDOP math and checking it for errors(I do kind of know this stuff).
As I suspect we may be seeing a bug in the APM firmware built for the iris, I will also use a utility to chat with the ublox directly and see if its telling me a different story than the pixhawk but NOT doing that until I see the swapout do the same thing , at this point I think we may be looking at a few flaky ublox chips(NOT unheard of from other vendors also).
HZL
ps you are very correct in being extremely cautious with the IRIS until this is resolved as it it so heavily dependent on a properly working GPS.
I don't think it is just an IRIS problem. I fly an APM board with a different GPS unit and have experienced the same issue. A flying buddy of mine flies NAZA and has the same issue. When I have an issue he has the issue. It will come down, but sometimes it takes 10 minutes or so. I'm guessing it has to do with what satellites are visible at the time, and how they are being moved around right now to accommodate some out of service ones, and also the new ones being launched. If a bunch of the satellites visible to you are near the horizon, the antenna on the ground will have a hard time picking them up. I also have notice when I am flying my full scale plane with the latest GPS software and hardware I have been getting degraded signal messages pretty frequently. If I go a few minutes and the HDOP is stuck and not dropping I unplug and restart the APM. Seems to change something and get it to lower the HDOP quicker. Also I have noticed once it has found itself subsequent restarts happen almost immediately. I am pretty certain its a GPS issue with the sats and not the actual IRIS. When I have witnessed 3 completely different systems in the same day, around the same times having the same issue using different antennas, different firmware, and different software. I started a thread about it more than a month ago but it got buried with lack of responses.
hmm but that doesnt explain why both iphone 4S and Android(GTP313 were getting 3 and 4 meter fixes in a basement with wifi turned off(I have loaded up utilities that allow me to see the satellite constellation numbers of satellites and precision of HDOP and VDOP and other navigation fixes)
while Ublox 6H in the IRIS with or without a copper ground plane is seeing same exact satellites BUT cant get HDOP below 2.1 rarely... if NOT defective GPS chip antenna then we are looking at some serious desensitization here or software errors in how HDOP is calculated.. hmm if HDOP is same issue on NAZA that would point to same code with same bug, does NAZA use the Ublox 6H by per chance? I am tracking accuracy of HDOP on 3 separate devices and logging and the IRIS using the ublox 6H seems to be having the issue.. not the Satellite constellation, so next after GPS swap for me is a utility to directly log the fixes coming from the Ublox 6H and log those along side the iphone 4S GPS sensor log and the same for the android gtp3113. Have to see if HDOP calculated in the APM code or in the UBLOX 6H directly(I suspect the latter).
Until my navspark chips arrive later this month(hopefully) I dont have a comparison independent GPS chip to test these suppositions against..(at this rate I will hook up a navspark in GLONASS mode for a sanity fix on these GPS issues.)
HZL
ps ordering 2 RAW mode navspark for GCS usage so can have accurate GCS fix and xmit corrections with software I have yet to develop
https://sites.google.com/site/wayneholder/self-driving-car---part/e...
seems to get a LOT more accurate fixes than I do from the exact same GPS but on a bread board..
This discussion here seems to touts the NEO 6M as getting 1.3 HDOP under a metal roof even though a cheaper GPS(one of those arriving monday from goodluckbuy)
http://diydrones.com/forum/topics/apm-2-5-with-neo-6m-ublox-gps-not...
its work a try to get things working.. even a much cheaper unit...
HZL
the second discussion with http://diydrones.com/forum/topic/listForContributor?user=3dsw9jz7wfl9y seem to point to a another solution.. slow down the update rates to 2-4 hz get 10-14 sats instead.. set mode for pedestrian on the 6H to improve loiter and prevent GPS wander he has a config file and APM code changes in that discussion.. I will be looking into the code he is pointing at in APM.
HZL
ps the MTK GPS is mentioned also as getting 10-14 sats and not getting bad HDOP
I get 1.3 routinely once it captures. Cell phones also cheat because they use the towers to get a quick position so they know the area they are in before they start looking for sats. I don't know much about the codes or refresh rate issues. I am not even close to a programmer and don't know squat about code. But once captured my gps works great.
actually the utility I m using is accessing the GPS device directly so no the cell towers DONT read into the utilities display of HDOP(which is coming from the GPS only.. cell tower access has no HDOP equivalent) and cant on the android device as it is only wifi and that turned off. My GPS is defective from what I can tell. am looking for a PIXHAWK passthru utility so I can use ucenter to chat with the GPS directly and need my new Micro USB hull port as that broke first thing to be delivered to do any of this(replacement GPS in same package).
HZL
-
1
-
2
of 2 Next