I've just uploaded the latest Ardustation 2 release - 2.0.15 which adds Mavlink 1.0 support. This is to support the upcoming releases of ACM and APM with Mavlink 1.0. You can still use this release with Mavlink 0.9 by commenting out the "#define MAVLINK10" statement in the main ardustation2.pde file.
This version will also warn you if it detects the wrong Mavlink heartbeat being transmitted. Ardustation 2.0.15 will display an error message on the LCD when it detects the wrong heartbeat and direct you recompile with the opposite state of the #define MAVLINK10 statement. The start up message displayed on the LCD will indicate if you have compiled for Mavlink 1.0 or Mavlink 0.9.
I've have tested Mavlink 1.0 with ACM 2.6 Gamma, APM 2.4 (with Mavlnk 1.0 enabled) and with the Mavlink 1.0 version of the Mission Planner.
As always, please let me kinow if you have any issues with this release.
email@example.com or comment on this forum post.
and is also available in my git clone here:
and you will need a git client to pull the clone down.
I'm trying to think of different possibilities. What is very interesting is that you mentioned you tried with a direct cable and no heartbeat was received. Was the 3DR radio completely removed from your direct cable? In otherwords, did you just hook up:
APM2 Serial TX <-> Ardustation Serial RX
APM2 Serial RX <-> Ardustation Serial TX
APM2 GND <-> Ardustation Gnd
Also - have you tried requesting parameters - since this should work even if the heartbeat is not received?
In any case - the next step would be to modify the Ardustation code or perhaps find a serial diagnostic that would display what bytes were received by the Ardustation from its serial port (if any). I'll see if I can come up with something that displays on the LCD the status of the serial port and what it sees. This will take me a couple of days to work out but I'll let you know when I have something for you to try.
yes I had connected APM2 and ArduStation in that way. The 3DR radio module was no longer connected. Flight data or parameters are not sent (have requested)
Can you imagine that there is a timing problem?
Thank you for your interest and your work. Joachim
The fact that the 3DR radio is out of the picture is a good thing because I would have originally said that the way the 3DR radio handles rf transmissions might be a problem with new logic I added to compensate for the Xbee dropping bytes during transmission. Without any rf technology in the loop means that it could still be the logic I added clobbering the heartbyte because of the unique timing of your APM2 and Ardustation. However I would have expected that to occur with 0.9 also - so it is definitely a mystery to me right now. I'll will create a 3 conductor cable and hook up my quad and Ardustation in the same way to take the xbee out of the picture and see what happens then. When I'm set up this way, I'll add some diagnostics to the code to display the values of the first 20 bytes of the 1HZ frame and basic timing of when they arrive so that I can identify if the heartbyte is being corrupted or tripping my parser abort logic. That will also say if serial comm is shut down completely on your setup for some reason. The first message transmitted in 0.9 or 1.0 Mavlink is the heartbeat, so it should be obvious once I add the debug to see why the parser doesn't see it or any other traffic for that matter. The fact that parameter download doesn't work for you tells me that the "start" message that turns on Mavlink 1.0 is not working. I bet you're not getting any traffic, even when you select start from the Ardustation.
I bet the Ardustation would see traffic if you also had the mission planner sending the "start Mavlink" message to the APM2 like I do when I have multiple ground Xbees running. This is difficult because you can't connect both the laptop running mission planner and Ardustation as easily as with multiple Xbees. I once had a bug where the start did not work, but I could start the traffic from the mission planner and things did work.
I'll setup my non xbee mode with my Ardustation and fool around with it to see what I can add to help you see if the traffic is occurring at all or not or if the heartbyte is being corrupted. I'll also look at the Mavlink 1.0 "start" command to see if something may be up there.
I just finished a debug version that will allow you to display data values and a count of bytes received from the Ardustation's serial port. I'm calling the version 2.0.16, but I don't want to distribute it from the normal distribution location for now. Please send an email to my email address (firstname.lastname@example.org) and I'll reply with the URL to download a zip of this version.
I've added a new top level menu item (select down until you see "Serial Stat" and then press the center button. It will display on the second line of the lcd the number of bytes received every 500 msecs. On mavlink 1.0 you should see about 320 bytes every second if "Start Feed" has been selected. If the feed is stopped, it should receive 17 bytes a second (just the heartbeat). On the third and fourth lines of the LCD will be displayed the hexadecimal values of 10 bytes received after half a second of no bytes received (essentially the start of a frame of Mavlink data). The heartbeats will be displayed as:
FE 09 XX 01 01 00 02 00 00 00 where XX is an incrementing sequence number.
If you run this version, I would be interested in the number of bytes received and if you see any bytes at all displayed.
I haven't had a chance to wire up a 3 conductor cable and will try to do that tomorrow night.
Hello to all.
Since the conversion to 2.6. I did not receive data from the APM2 to Ardustation with Heino's software 2.0.15.
Heino has helped me a lot. Since I do not have appropriate measuring means, he made a program written for Ardustation, indicating the received information to the display of Ardustation.
Also, a program to test the Rx and Tx functions he has written for me.
Ultimately, everything works again and I'm interested here in the forum once again to thank Heino for his effort and work.
Now it is clear that Mavlink 1.0 also works fine on 3DR radio between APM2, PC and ArduStation.
Heino, you've done a great job!
Sorry for my bad english!
Yes - this software is for the $59 ardustation. As far as a tracker antenna stand - the one I still use is the readymaderc unit. It is definitely not the most robust patch antenna mount and the servos will wear out with time, but I've used it now for over a year with the same servos and it works fine for my purpose.
This is the unit I'm using:
Here is a picture of my setup:
And the video showing it running:
There are a number of more robust systems that Ardustation users have built by themselves, or purchased, but I still use the readymaderc unit.
Thanks a lot for the update and also keeping your git clone up to date. Makes merging changes so much easier. I didn't get as much flight time this summer yet at least not as much as I would have hoped for. Since I switched APM to a different plane I'm still spending time tuning the PID levels.
Middle term I also plan to throw together an antenna tracker for a video link. Latest then I should figure out how you solved this but I guess I should just look at the code. Was planning to throw the tracker together with some servos, hot glue and shapelock, let's see how that goes.
Hi Heino...it´s possible to add a menu to set home altitude for the arduplane... similar to switch between absolute and relative altitud...this function it´s available in the mission planner...
Thanks for your attention..
It would be possible except that the software id very low on RAM on an ATMEGA328. I have a new version 2.0.16 that added diagnostic information for serial port debug and that broke the RAM limit until I made some internal variable changes. I haven't released it yet since I want to more extensively test it for side effects. I do store the station altitude when the home position is set, but displaying absolute vs relative would require some kind of display switch. I'll look at adding something like that but it may not be possible without removing some existing functionality.
Hi ... Thanks for your attention ... I have another question .... my tracker uses Futaba servos ... and when I did the test, the servo was beyond limits .. how I can add a limit on the software ..? ? the problem is below 50 and above 140 in servo angles ... I understand the servos futaba only have 60 degrees of movement on each side .... sorry for the inconvenience