Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

AC3.5: Pixracer params can't be updated over wifi #5799

Closed
rmackay9 opened this issue Mar 2, 2017 · 11 comments
Closed

AC3.5: Pixracer params can't be updated over wifi #5799

rmackay9 opened this issue Mar 2, 2017 · 11 comments
Labels
Milestone

Comments

@rmackay9
Copy link
Contributor

rmackay9 commented Mar 2, 2017

If using a Pixracer with the little wifi board, changes to parameters (i.e. ARMING_CHECK) always fail. This was tested on master (pre AC3.5-rc2) using windows PC and mission planner.

@rmackay9 rmackay9 added this to the AC 3.5.0 milestone Mar 2, 2017
@rmackay9
Copy link
Contributor Author

rmackay9 commented Mar 2, 2017

First reported by artemen on this thread: http://discuss.ardupilot.org/t/copter-3-5-rc1-released-for-beta-testing

@WickedShell
Copy link
Contributor

@rmackay9 This is the same underlying issue as #5505 I think. It's not something master plane suffers from. If you wait long enough you can download params quickly, but its still very hit or miss.

@rmackay9
Copy link
Contributor Author

rmackay9 commented Mar 2, 2017

@WickedShell, ok thanks. If it's the same issue then the bit of new info I have is that the performance is worse between AC3.4.5 and AC3.5-rc1.

@OXINARF OXINARF added the Copter label Mar 2, 2017
@WickedShell
Copy link
Contributor

I believe @DonLakeFlyer was saying it fell apart around the time of the frame change? I could be really off on that recollection however.

@OXINARF
Copy link
Member

OXINARF commented Mar 2, 2017

As I've written in discuss, the wifi port is just like another serial port from an ArduPilot perspective. If this is an issue, is it happening with other devices? Here is a link for a long discussion where @peterbarker was testing new firmware for the Wifi board and got much better results.

Updating to the most recent MAVESP8266 firmware has solved the vast majority of the lost packets on the downlink - I'm at 1.2% packet loss.

Looking at it, it makes sense; prior to this change a lot of the packets the vehicle sent would not have been recognized - ArduPilot support is actually a late-comer to the ESP8266 firmware. This is a drawback in having the ESP8266 interpreting the mavlink messages.

I don't know why it has gone worse with 3.5/master but it's possible we are sending more messages and the Wifi board just can't keep up.

@WickedShell
Copy link
Contributor

@OXINARF Plane (at least as of when I commented on #5505) wasn't suffering any problem like that.

@OXINARF
Copy link
Member

OXINARF commented Mar 2, 2017

This should be fairly easy to test:

  • connect a radio to the first telem port in Pixracer, does it work?
  • connect the wifi board to the same telem port, does it work?
  • connect the radio to the wifi port, does it work?

Don't forget to change the baudrates accordingly. If only the Wifi board doesn't work, it is almost certain that the issue isn't with ArduPilot.

@rmackay9
Copy link
Contributor Author

rmackay9 commented Mar 3, 2017

A bit more information:

  • the issue affects Plane (master) as well as Copter (and probably all vehicles)
  • the GCS->vehicle link is likely the cause of the slow parameter downloads (and inability to set parameter values) with many messages from the GCS not successfully getting through to the autopilot.

@lvale
Copy link
Member

lvale commented Mar 3, 2017

@rmackay9 I've been tracking this with Don here mavlink/qgroundcontrol#4643

still have to check via wifi, which will do after work

@peterbarker
Copy link
Contributor

If you could try 026c65d3646b0f5671bf5c5a7d58c5d2cc82844c then try 026c65d3646b0f5671bf5c5a7d58c5d2cc82844c^ to see if you see the same thing that would be great.

Note you'll want a git submodule update between as it is a submodule commit in question.

@peterbarker
Copy link
Contributor

This problem appears to be resolved.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

5 participants