Georacer
October 16, 2024, 9:34am
1
Attendees (max) : 9
ArduPilot:master ← Georacer:pr/wiggle-tecs-reset
opened 03:09PM - 13 Aug 24 UTC
For the balloon drop use-case, I've noticed that TECS would initialize on the gr… ound. That meant that the desired altitude (which is under an LPF) would not start close to the actual altitude, rendering itself useless for an RTL:
With this commit TECS starts at the release altitude, resulting in reasonable altitude error:
:warning: Requires https://github.com/ArduPilot/ardupilot/pull/27758 .
UTC0757
Merged!
ArduPilot:master ← Georacer:pr/takeoff_idle_throttle
opened 07:12PM - 08 Oct 24 UTC
I thought I'd tackle https://github.com/ArduPilot/ardupilot/issues/20754.
A n… ew parameter `TKOFF_THR_IDLE` has been added and a new option bit 1 to `TKOFF_OPTIONS` to enable it (off by default).
In TAKEOFF and AUTO modes, while in `FlightStage::TAKEOFF` and while the throttle is suppressed, `TKOFF_THR_IDLE` is output to the throttle.
Here's how it looks like in a catapult takeoff, for `TKOFF_THR_IDLE=20`:
UTC0759
Andrew: is_zero(param :get()) should workGeorge : Ok, I’ll try again.
ArduPilot:master ← Georacer:pr/quadplane_refactor
opened 10:26AM - 11 Oct 24 UTC
According to @peterbarker 's suggestions here: https://github.com/ArduPilot/ardu… pilot/pull/28329#discussion_r1793096558
UTC0802
Merged!
ArduPilot:master ← haydendonald:pr-add-mcu-voltage
opened 10:35PM - 07 Oct 24 UTC
Add analog:mcu_voltage() to get a reading of the mcu voltage
Tested on CubeOr… angePlus
UTC0803
Merged!
ArduPilot:master ← bugobliterator:pr-periph-imu
opened 06:03AM - 13 Oct 24 UTC
* Also fixed build with both scripting and GCS enabled
* Changes name from Cube… Pilot-CANMod and -PPPGW to CubeNode
UTC0805
A : can_imu_update() could use a small sleep.Sid : Do but we don’t do that on the vehicles?A : But the main thread here doesn’t do anything more important. So this could turn into a hard-spinning thread.S : I’ll add 100us, it’s fine.A : There’s one change in the bootloader file. Let’s make sure that only the CubeNode bootloader gets affected.
I’m not sure about the changes in the .h files. Peter should check them.
UTC0902Peter : Will need some cleanup on the build server directories.
I’m using Tools/scripts/size_compare_brancyes.py to compare the binary size changes.
No changes on the vehicle bootloader binaries.
Left change requests.
ArduPilot:master ← bugobliterator:pr-fix-ccache
opened 03:37AM - 16 Oct 24 UTC
Following error is shown when not including path:
```
[2024-10-16T11:03:34.106… 740 1906419] Detected input file: ../../libraries/AP_Filesystem/AP_Filesystem_FATFS.cpp
[2024-10-16T11:03:34.106803 1906419] Source file: ../../libraries/AP_Filesystem/AP_Filesystem_FATFS.cpp
[2024-10-16T11:03:34.106804 1906419] Dependency file: libraries/AP_Filesystem/AP_Filesystem_FATFS.cpp.0.d
[2024-10-16T11:03:34.106805 1906419] Object file: libraries/AP_Filesystem/AP_Filesystem_FATFS.cpp.0.o
[2024-10-16T11:03:34.108073 1906419] Trying direct lookup
[2024-10-16T11:03:34.108127 1906419] Failed to lstat nano.specs: No such file or directory
[2024-10-16T11:03:34.108128 1906419] While processing --specs=nano.specs: nano.specs is missing
[2024-10-16T11:03:34.108140 1906419] Failed; falling back to running the real compiler
```
I believe this happens when toolchain is not installed in system folders. In any case I believe we should specify the path to the specs, to avoid ccache doing false match with what it suspects are the specs being used vs what actually will be used.
UTC0817
S : This fixes things on my system. Build doesn’t work right now.A : Doesn’t look right.
You could try the --ignore-options flag to get the error to go away.
S : But it doesn’t fix my issue.A : You could try to strace the error and find out what the actual problem is.
ArduPilot:master ← bugobliterator:pr-spi-gpio-duplicate
opened 10:08AM - 15 Oct 24 UTC
This fixes a bug where we run GPIO loop through SPI two times per pin as we dupl… icate MISO and MOSI name with RX and TX as well. So if we have GPIO(x) next to SPI the hwdef fails to compile, and does not allow SPI lines to also be treated as GPIO.
UTC0826
MergedOnCIPass.
ArduPilot:master ← bugobliterator:pr-spi-perf
opened 06:48AM - 13 Sep 24 UTC
* Tested on CubeOrangePlus and CubeRedPrimary with Invensensev3 driver
* I us… ed debug trace to analyse timeline for read_fifo to check if we have any substantial overhead while doing SPI transactions. It turned out that calling spiStart and spiStop which in turn call dmaStreamAlloc and and dealloc to take substantial amount of CPU call almost 3-4us. Also we were calling Shared DMA mutex lock and unlock when the DMA is not shared between peripherals.
* Following changes made SPI transactions much less CPU intensive:
- * Do not reset SPI peripheral using spistop and spistart calls if SPI config hasn't changed.
- * Also make changes to SPI frequencies for CubeOrangePlus and CubeRedPrimary to match on the bus where possible (minor CPU gain through this, just first part was enough as most transactions are IMU anyways).
- * Do not call mutex lock unlock for unshared DMAs.
Results:
CubeRedPrimary:
Before:
SPI1 PRI=181 sp=0x20001AA8 STACK= 792/1464 LOAD=13.4%
SPI4 PRI=181 sp=0x200021C0 STACK= 872/1464 LOAD=13.4%
SPI2 PRI=181 sp=0x20005380 STACK= 896/1464 LOAD=13.3%
After:
SPI1 PRI=181 sp=0x20001B08 STACK= 792/1464 LOAD= 9.7%
SPI4 PRI=181 sp=0x20002220 STACK= 792/1464 LOAD=11.2%
SPI2 PRI=181 sp=0x200053E0 STACK= 840/1464 LOAD= 8.3%
CubeOrangePlus:
Before:
SPI4 PRI=181 sp=0x3002AF58 STACK= 668/1464 LOAD=25.5%
SPI1 PRI=181 sp=0x3002B668 STACK= 668/1464 LOAD=14.5%
After:
SPI4 PRI=181 sp=0x3002AF78 STACK= 668/1464 LOAD=19.2%
SPI1 PRI=181 sp=0x3002B688 STACK= 668/1464 LOAD= 9.5%
>10% total CPU time recovered for both CubeOrangePlus and CubeRedPrimary
UTC0827
A : Needs some changes to ensure that the heap does not overflow.
ArduPilot:master ← tridge:pr-external-AHRS-variances
opened 03:25AM - 15 Oct 24 UTC
simplifies backends, allows GCS to see variances in EKF message, and allows for … copter EKF failsafe to work
broken out of #24747
UTC0840
A : This will allow proper support for external AHRSs, because it will allow for EKF lane switches, in case of solution degradation.
ArduPilot:master ← tridge:pr-AHRS_CINS
opened 04:51AM - 24 Aug 23 UTC
This is the ANU CINS estimator worked on by @patwilt and @pvangoor
This is stil… l experimental but is giving promising results in test flights
remaining issues:
- [x] using std::queue, should use ArduPilot queue APIs
- [ ] should support baro for height for handling GPS outage
- [ ] add option to not use compass
UTC0845
A : Will need a reviewer.
Then we’ll need a lot of testing to make sure it doesn’t break in action.
The idea is for this to eventually replace DCM.
ArduPilot:master ← peterbarker:pr/ap-bitmask-stop-it
opened 01:09AM - 09 Oct 24 UTC
Table edited for new iterate-over-cast-structure scheme:
```
CubeOrange-periph… -heavy *
CubeRedPrimary * * -448 -408 * * *
Durandal * * -272 -128 * * *
Hitec-Airspeed *
KakuteH7-bdshot * * -424 -360 * * *
MatekF405 * * -376 -352 * * *
Pixhawk1-1M-bdshot * -416 -360 * * *
f103-QiotekPeriph *
f303-Universal *
iomcu *
revo-mini * * -384 -352 * * *
skyviper-v2450 -464
```
Tested in SITL by generating a simple log-disarmed log and looking at the message, and using `du32_change.py` to see bits are changing appropriately
UTC0848
Randy : If Pete wants to merge this, feel free.
ArduPilot:master ← peterbarker:pr/rangefinder-over-327
opened 05:57AM - 21 Aug 24 UTC
UTC0852
A : It will have to wait for 4.7, unfortunately.Pete : That’s fine.
ArduPilot:master ← robertlong13:pr/flightaxis_improvements
opened 09:36AM - 02 Oct 24 UTC
Adds a new options parameter for the FlightAxis SITL frame:
- Option to automati… cally reset the aircraft position in RealFlight!!
- Move the existing framestring-passed options to option bits (but keep the existing framestring parsing for now)
- Option to silence the FPS reporting printf
UTC0929
MergeOnCIPass
ArduPilot:master ← tridge:pr-q-guided-terrain-fix
opened 08:19AM - 15 Oct 24 UTC
if the GCS commands a guided re-position with an AGL frame and quadplane is in Q… _GUIDED_MODE=1 then this prevents us adding the home altitude to the target terrain altitude
we really need auto tests for these
UTC0855
A : This is pretty urgent, leads to a large climb.
Needs autotests before it gets merged.
There will be NO Dev Call on Tuesday!