Hello,
I am building a large skid steering rover using Cube Orange+, ADSB carrier board, Here4 GPS, RFD 900x and Mission planner for GCS. I am using rover 4.5.1 and having issues with mode changes. It says “Flight Mode Change Failed” this happens when trying to switch out of hold and into auto or RTL. This issue is also found when testing with the Simulator.
I have 3d fix (20+ sats with about 0.6 HDOP), no PreArm checks, Just completed Accel Calibration and compass calibration.
The attached Log File is with the carrier board and Here4 attached to a box outside in order to isolate any possible interference problems. I also flashed the board back to very basic params to make sure that I hadn’t unknowingly changed a parameter somewhere.
Any ideas? I"m about to scrap the project if I can’t get past this.
Have you uploaded a waypoint mission?
Have you armed it?
If the answer to either is no, the results you’ve observed are by design.
It seems you are putting the cart way ahead of the horse by attempting autonomous navigation with an autopilot that has barely been configured, much less installed in a vehicle and tuned.
It was fully installed in the vehicle, had this issue so I went ahead and pulled it so that I could do calibrations. (My vehicle weighs approximately 500lbs)
If it won’t work on the bench, it won’t work in the vehicle.
You have no RC receiver configured, which is causing a failsafe.
You have a safety switch enabled (BRD_SAFETY_DEFLT=1), which will also preclude output if not installed and activated.
You have all arming checks disabled, which will indeed allow you to arm, but may mask some of your underlying issues.
There are some EKF messages that seem out of place, so initial calibration steps may need to be re-accomplished.
Even if you do not intend to use an RC transmitter during routine operation, you really ought to configure one for initial tuning and setup. Attempting auto mode on a large vehicle without having driven it manually and accomplishing at least some rudimentary tuning steps is asking for problems (or injury).
Your uploaded mission crosses a public road, a treeline, and a driveway, at a minimum. It’s probably a good thing it didn’t work.
I may not have been clear enough. The vehicle has an Rc that has been configured and driven manually with no issues it was when changing modes that the problems occurred therefore causing me to pull the board and gps to test on the bench.
I’ll try to get a log uploaded from the previous full vehicle assembly tomorrow.
This mission was a 3 second write going on a cardboard box, that wasn’t going anywhere, to see if it would even attempt a mode change. I am very familiar with writing missions and would never have done this if there was a chance that it could execute it.