Plane-4.7.0-beta2 has been released for beta testing and can be downloaded using Mission Planner, QGC or directly downloaded from firmware.ardupilot.org.
With the 4.7.0 beta 2 this morning, under beautiful skies, I flew fixed-wing and VTOL without any issues using FBWA, AUTO (mission created from LUA script), CRUISE, GUIDED, TRANSITIONS, QSTABILIZE, and QHOVER.
I’ve been flying dev builds of 4.7 on a couple of planes for a while now, because they have F405 Navi Deluxe boards in them with no earlier target. I recently set up another plane with 4.7.0 beta1 and flew it, then updated it and one of the others with 4.7.0 beta2. All of them that were set up and tuned in 4.7 are flying really nice! I’m not sure if changes were made to autotune since the last time I did one, but I’m quite happy with these 3 planes so far. I can confirm the throttle slew rate fix in beta2 (PR:32381) is working as intended. I wondered before why the throttle would often slam immediately to the set takeoff throttle, and occasionally would ram up according to the set slew rate. Now I know! lol
Thanks to everyone who continues to make Ardupilot as good as it is!
I think I may (or may not) have found a bug… For context, my plane has the following:
AR Wing Pro
Matek F405-Wing (original)
Arduplane 4.7.0-beta2
Qiotek ASP5033 airspeed sensor
I’ve flown it twice on 4.7.0-beta2, on the first flight I noticed that after taking off (using takeoff flight mode) while loitering overhead, I could hear the throttle pulsing. In the FPV feed I also noticed the temperature reading on the OSD (taken from the airspeed sensor) was bouncing around all over. I switched to FBWA and flew for a bit, and saw the temperature seemed to be normal, so I switched to cruise, and it was managing throttle as expected at that point, and for the rest of the flight.
Before the next flight, I set ARSPD_USE = 0 so it wouldn’t use the sensor for navigation, but would still log it. I also checked the tubing and pitot for blockages, leaks, etc and it all seemed good. I flew again, and saw the temperature fluctuating at the start of the flight like before, then it returned to normal again.
For 5033 this happens if no data can be read for more than 0.1 of a second. Could be a i2c wiring or power problem. I don’t think it will be related to the beta firmware.
I assumed it was a problem with I2C data, the logs show a clear difference in the rate at which I2C interrupts are processed that matches with the sensor being weird, they returning to normal.
I will double check wiring for any problems that might have developed there, but I do find it somewhat telling that the logs from both flights looks very similar, starts out acting up then becomes “normal” shortly into the flight, and is good for the remainder. I’ve got a number of flights on the same plane with no physical changes, and it just started after updating the firmware. I don’t remember the firmware version I updated from, but it was likely 4.4 or earlier. It had been a while since I flew this one until recently.
I will try to narrow things down, and see if I can repeat the issue on the ground. If so that will make things a lot easier to diagnose… lol
