RTK Loiter / RTL Testing with AC 3.2.1 and NS-HP-5

Holidays Greetings!

There is much interest in seeing demo of our $50 pre-order NS-HP RTK receiver on a quadcopter, so we tested last week on a less windy day (with up to 4.5mph wind). A compass module and 5Hz version of NS-HP RTK receiver replaced the original GPS and compass inside the 3DR Iris. These two videos show the result we obtained with default values without any tuning.

E-mail me when people leave their comments –

You need to be a member of diydrones to add comments!

Join diydrones


  • Ok, have just make our first test.

    Everything seems to work fine with latest firmware, got float RTK solution in less than a minute (unfortunately, didn't get Fix ten minutes later).

    BUT working with NSHP in ROVER-FLOAT MODE instead of Rover-Normal mode, Gnssviewer didn't announced to get into FixRTK EVEN with RTK-Ratio over 3 (we got proven 2cm accuracy in viewer), thats odd. When moved to Normal mode willing to show the fix announced in viewer we couldn`t. But that was only our first test. Satisfactory.

    NSHP with HC-05 bluetooth worked as well, linked to an android phone with Lefebure apk.
    NTRIP casting over bluetooth worked (didn't get rtkfix, but data transmission worked showing 1 yard accuracy).

    Very happy with those units in our first testing environment.

    Keep working in this system guys, you are doing it fine.

  • There are some earlier discussions on how to trigger camera with Pixhawk:



    The same Pixhawk trigger signal can trigger NS-HP-5S to generate RTK-accurate position via STI,005 message.


    Today we just released a Windows-based RTK post processing utility for NS-HP-5S users that wish to log raw measurements and post process to generate precise camera trigger positions without needing to setup radio link between base and rover on the UAV. It takes base and rover raw measurement files and generates a file with RTK-accurate camera trigger positions (similar to STI,005 message), and another file with measured positions at regular intervals (similar to RTK NMEA output). The utility is available on NS-HP product page; section 10 of the updated user guide has description on this.

  • Any news? I would like to use this for mapping, is there a way to sync with a sony camera?

  • We have just received 2 units (base-station) plus interesting freebies: a micro uart-to-usb dongle wich I think could be embedded in the gps shield, and a fancy minimal gps development board, size of a nail LOL.

    Really interesting units and very fast delivery. Still can't believe these are a L1 gps complete system  ;)

    We'll test'em soon.

  • Dear Oliver, have just ordered two of your 5hz units for testing.

    Do you have NTRIP communications documented? (just asking, not needed by now)


  • Raw measurement output is in SkyTraq binary format: http://navspark.mybigcommerce.com/content/AN0030_1.4.31.pdf

  • Hello, could you tell me what the output format of the raw measurements is? Is it the proprietary Skytraq raw or RTCM 3.x format? If it is the Skytraq format, is there a protocol definition somewhere that might facilitate translation to RTCM 3.x? Thanks!

  • We don't see higher update rate any better than lower update rate. It should mainly be dictated by application requirement. Arducopter code default requires 5Hz GPS, thus using 1Hz RTK the behavior appears to be a bit strange, jerking sometimes. 

    It's as described in the user's guide, the only configuration needed is if setting up a local base, using Windows GNSS Viewer program to set into base mode and configuring base antenna position. 

    Using telemetry radio link for sending base data should be no problem for everyone. Setting up NTRIP accessible base is more complex, same complexity as using RTK receiver from others, thus this topic is not described in NS-HP user's guide. 

    NS-HP with -S time/position stamp option can output exact time & position of the receiver antenna at instant of the rising edge trigger signal on TRIG input pin. It is output in proprietary NMEA PSTI,005 message in NMEA output stream and in binary form over TX3 raw measurement output stream. This mainly might be useful for aerial survey mapping applications. 3:36 ~ 4:03 of this video explains advantage of this approach.

    The position stamp is highly accurate to 1 millisecond traveled distance on top of 1cm + 1ppm RTK error; i.e. for 50kph UAV, less than 3cm error when operating in 5km baseline (50000m/3600/1000*100+1cm+1mmx5). A 10Hz RTK without such position stamp, travels 139cm between these 10Hz position points. it would be difficult to estimate position to centimeter-level by precise time stamp if UAV is not moving at a constant velocity in a straight line.

    With this GPS-based equal-distance taking an image approach, one could only take an image when UAV comes within say 5 meters of the planned flight path waypoints. Position stamp gives pretty accurate position (within 3cm error bound flying 50kph within 5km baseline for example) no matter where the UAV flies by within 5m radius of the waypoint. Much more difficult to achieve same level of accuracy if using 10Hz RTK without position stamp function.

  • Dear Oliver we are going to try your NS-HP series (next month) problem is -as every new development- the absolute lack of information out there. Some questions here:

    Performance in 10Hz unit is noticeable over 5Hz units?

    True plug and play following your instructions, both in rover and Base Station? (no programming, more than rtklib)

    What does mean Time and position stamp? is refered to data offered over NMEA?


  • In the earlier two replies to Nikola, one can see after getting RTK fix, there should be no jump in position output to cause drone to fly away suddenly if the interval dropping to float is short 90 seconds for example. If loosing RTK fix for a very long time and the receiver decides needing to restart the RTK process, then there will be jumps. 

This reply was deleted.