@ChrisDaily wrote:
We have a small group of us here at the shop who are all drone guys. Must of us are fairly versed on Betaflight and DJI, but could use some help with Ardu. Anyone in the Sherman or Dallas TX areas willing to help our group out? If so feel free to give us a call at 903-819-1452 or email at: chris@texomaclassics.com
Posts: 1
Participants: 1
@markb2309 wrote:
Hello,
Does someone have the v0.2.2 firmware for the herelink airunit
the download page on cubepilot doesn’t work so i can’t download the firmware
Posts: 1
Participants: 1
@rmackay9 wrote:
Here are the slides of the monthly report for Mar 2020 presented by Tridge, Randy and a few other developers at the monthly Partners call.By the way, the image on the left is a DIY test bench for Andy Piper’s work on the FFT filtering. The image on the right is from this video demonstrating Paul Riseborough’s progress on enhancing the EKF so that it can automatically detect the vehicle’s yaw using a Gaussian Sum Filter (GSF) without requiring a compass.
As always, thanks very much to our Partners for their support!
Posts: 1
Participants: 1
@arduouspilot wrote:
I would like to use the momentary switch (SH on Taranis X9D Plus) to set the flight mode to RTL. Now, I thought I had done that but it appears to not work.
It’s probably best just to look at my screenshots:
Aim is to get logical switch L07 set from a momentary flick of the spring-loaded switch SH. I then make L07 cause the appropriate special function to set RTL mode. Note that SF1-6 work fine and as expected.
When I flick SH, nothing happens.
Posts: 1
Participants: 1
@rmackay9 wrote:
This is a demonstration of ArduPilot using an Intel RealSense T265 camera for non-GPS navigation using the scripts created by Thien Nguyen (see blogs here) as part of GSoC 2019 under the wise mentorship of Patrick Poirier.A few others have already successfully reproduced the setup but I wanted to use it myself for an upcoming search & rescue competition which requires the vehicle to fly into a building. I also wanted to make sure the setup was easy for others so Jaime Machuca and I created an APSync image for the RPI4 (details are here on the wiki).
Overall I was quite surprised at how well it worked. You can see especially at the beginning of the video that the vehicle’s position hold is very stable, more stable than when using a GPS actually. Also about halfway through the video I tried flying it higher with no objects near the vehicle and it still performed quite well.
On the other hand if you look at the end of the video you’ll see that it completely loses its position estimate after I yaw it around a few times. In the image below the yellow line shows where the EKF thought the vehicle was (over in my neighbour’s yard) but in reality the vehicle was always safely in my backyard.
I also bumped into a known but very annoying hardware problem with the T265 which requires the camera to be powered up after the companion computer. This is resolved in the APSync image using uhubctl to power-down/power-up the USB ports after the RPI4 has started.There is more work to do on the integration of this camera with ArduPilot but it already works well. If you want to get involved I recommend getting a vehicle working yourself with the T265 (wiki) and report any issues you find. If you’re a developer looking to help us advance the non-GPS features of AP then ping us on the ArduPilot/VisionProjects gitter channel.
By the way, the frame is an Hexsoon EDU450 with a CubeOrange autopilot.
Thanks again to Thien, Patrick and Jaime for their help with this.
Posts: 2
Participants: 2
@benglish wrote:
I’d like to be able to hook up more servos and motors than the default/aux output channels allow. Is there a way to convert some of the unused UART (or other?) channels into PWM or DSHOT outputs?
Reason:
I’ve modified an eflight convergence using a Pixhawk 4 mini running a custom flight controller and it was epic, but now I’d like to use a smaller flight controller to fly eFlight’s V-22. Most are limited in terms of outputs and I’d love to figure out a way to have more outputs for this and other aircraft.
Posts: 2
Participants: 2
@Rusty_Jones wrote:
Not sure what is wrong here, maybe the way I am doing this. Just uploaded AT 1.1.0 to an old generic “Pixhawk” Not sure where I got this one, but did have it sort of working on Tracker about a year ago. I wanted to update to current firmware, and now I can’t calibrate the accels or the compass. When I press the button to calibrate the accels it just changes the button to “Click when done” I remember it asking you to place on all six sides then press space , did that change?
Posts: 3
Participants: 2
@Avaneesh_Maithil wrote:
The Problem which I am facing right now is after arming the system after certain period it topples itself. I am not using any RC with it. I was using mavproxy and dronekit to fly the system. I don’t know why it happened So, any guidance will be appreciated.
Posts: 2
Participants: 1
@VDLJu wrote:
I have posted the issue to ArduPilot Github, but so far there hasn’t be any discussion yet.
See the issue for more detailed information: https://github.com/ArduPilot/ardupilot/issues/13875In my opinion the main issue here is that it is not clear if the current way of the execution is purposefully like that. This makes it harder to predict the behavior on mission execution as it may be a sum of multiple commands in certain cases.
This makes the it hard for 3rd party mission planning software to do abstraction as the small details are so important on certain combination of DO and NAV commands and how they are ordered. I’m hoping that at some point developers could define a “standard” or guide a how missions must be executed by autopilot. For example some kind of simple state machine that would make it easy to replicate and check generated missions on 3rd party planning software.
Posts: 1
Participants: 1
@Wing_Kuan wrote:
I am new to diy drone. My drone does not have Transmitter and Receiver. It has Arducopter v2.8, hexacopter and a GPS module with compass.
I am a mac OS user, I installed APM Planner 2.0 and found a way to upload Python Script which is install “MAVProxy” by using terminal. However, I can’t upload the script to the APM to execute it.
Terminal installation:
sudo pip install mavproxy
sudo pip install dronekitI tried to have two terminals.
First is with code “mavproxy.py --master=/dev/tty.usbmodem14201 --baudrate=115200” to read the drone.Second is “cd /Users/name/Desktop” then “sudo python takeoff_and_land.py --connect udp:127.0.0.1:14550”[https://gist.github.com/dbaldwin/9185b702091148580fa836c1911f8735#file-takeoff_and_land-py]
(Attached results below)version of lists:
dronekit 2.9.2
dronekit-sitl 3.3.0
pymavlink 2.4.6
MAVProxy 1.8.18Hope to get the answer (correct coding/steps) to execute the script without the RC receiver and transmitter in APM Planner 2.0.
Thank you in advance!!
Screenshot 2563-03-25 at 10.21.56 PM|535x500
Posts: 1
Participants: 1
@robodaniel wrote:
Hello,
I’m using a Holybro PX4 and Herelink HD Transmission system.
On the PX4 is running ardupilot but the herelink ari unit doesn’t seems to be booting. I’ve connected the Air unit to the SBUS RC and Telem1. In Missionplanner I’ve configured Serial1 as Mavlink (1) with a Baudrate of 57 000 but the LED’s of the Air unit are still off.Does anyone else had the same issue and knows how to slove it?
Thanks in advance
Posts: 1
Participants: 1
@dickou65 wrote:
Please help, MP 1.3.70, video stream display on HUD but all filght data, altitude,air speed, battery voltage etc. all disapear. all this flight data display on HUD before but disapear suddenly. anyone know what wrong?
Posts: 1
Participants: 1
@danielh wrote:
Hi,
I’m flying in guided mode with an optical flow (pmw3901) and lidar.
My companion computer sends set_position_target_local_ned
with velocity and yaw rate fields.
I flew several times very successfully and then for some reason the controller started to ignore the yaw rate field, while the velocity was as required.
Any idea why this might happen?
Unfortunately, the log (GUID) doesn’t show the yaw rate field, it only shows position and velocity. Is there a way to log the commanded yaw rate?Thanks!
Posts: 1
Participants: 1
@kelemvor wrote:
I’m new to remote control gadgets (at a technical level anyway) so I’m likely making a mistake. Hopefully someone can point it out.
My problem is that I can’t get flight controllers to “see” remote control input. At this point I’ve replaced my flight controller, transmitter, and receiver.
Originally I started with
Flight controller: Pixhawk 2.4.8
Firmware: PX4
Receiver: Tactic TRX600 (very old 72mhz fm receiver) with generic PPM encoder
Transmitter: Tactic TTX400
Now I’m using:
Flight Controller: Holybro Durandal
Firmware: ArudCopter 4.0.3 (ChibiOS)
Receiver: Spektrum AR620
Transmitter: Spektrum Dx6eHowever, after flashing that flight controller over to Arducopter 4.x I got no inputs from the PPM encoder. I posted separately about that and kind of decided maybe it was just too old to work.
I went to the local hobby shop and bought the best remote they had in stock, it was a Spektrum Dx6e which came with a Spektrum AR620 receiver. The receiver supports DSM2 and DSMX. 6 channels and it will do 11ms or 22ms. The transmitter only supports 22ms but still either DSM2 or DSMX.
According to the release notes, arducopter has supported dsmx 22ms since January so all should be well.
The transmitter and reciever are bound to each other. The indicator light and the data on transmitter’s display show that all is well.
I couldn’t get the Spektrum transmitter/receiver to work at all with the pixhawk flight controller running aruducopter. Neither via DSMx nor using the PPM encoder. When connected via DSM, QGroundControl shows there are no detected radio channels. When connected via PPM encoder, the channels show up but there is no response when I move any of the controls.
I was only connecting the DSM/data port (vcc, gnd, data pins) to the pixhawk so I thought maybe I needed to supply power to the battery connector on the receiver. Spektrum says it likes 3.5-9v so I went with 9v. That was a mistake as the battery port’s VCC is connected to the VCC wire on the DSM port… 9v fried a couple of the chips on my pixhawk.
So I bought a durandal flight controller from holybro. Loaded up Arducopter 4 on it, and… have the exact same problem!
At this point I’ve tried two different FC’s, two different receivers, and transmitters. I can make the hardware work with PX4 firmware but it just won’t work with Arducopter at all.
Here’s a photo of my wiring:
Any help would be greatly appreciated. Seems like I must be missing something simple. I’m assuming there’s someone in the world successfully using arducopter 4, so it must be something I’m doing wrong.
Posts: 4
Participants: 2
@bmag wrote:
I am building a quadcopter and plan to use a pixhawk autopilot with Ardupilot firmware. My understanding is that the attitude control law consists of an angular rate loop with PID control and a angular position loop with proportional control. I would like to set the PID by analysis, rather than using the AutoTune feature. I have a simple model for the pitch axis, but am having trouble designing a set of PID gains in the pitch rate loop that fit within the ranges given in the ATC parameter list. I suspect that the Ardupilot PID gains are scaled differently than my model. In my model the input to the PID control law is a pitch rate command in rad/sec and the output is a rotor speed command in rad/sec. My P gain is therefore unitless. I suspect that the Ardupilot P gain may not be unitless and may have units of rpm/deg/sec. Can anyone confirm this or tell me what units are associated with the Ardupilot P gain?
Posts: 1
Participants: 1
@smartdave wrote:
First and Foremost this is NOT bashing any Hardware at all. This is simply a question that I have.
I have a Tarot 690 Hex and I recently changed my CUAV V5+ running 4010Pro motors.
Since the fmuV5 only supports 4 Dshot outputs I decided to change the flight controller to a Cube Orange to take advantage of the 6 Dhot outputs.
However I noticed something I didn’t expect.
Nothing was changed except the Flight controller.
Here is the Vibes from the CUAV v5+
And here are the Vibes from the Cube Orange.
Again I am not bashing any product, I am simply trying to do some research.
Is this just a case of different isolation material? Or are the IMUs in the Cube Orange way more sensitive?
Posts: 1
Participants: 1
@KimHyungSub wrote:
Hello folks,
Default throttle value issue
I have a question about default throttle value (RC 3 channel) on SITL.
When I execute ArduCopter on SITL, it uses 1,000 as default throttle value as shown in the below.
On the other hand, other RC channels (roll, pitch, and yaw) use 1500 as default values.
Is there any specific reason to assign 1,000 to RC 3 channel? (i.e., why SITL does not assign mid-throttle value such as 1,500?)
This is because the drone loses his altitude if I am not assign a value more than 1,500 to RC 3.
Also, many guys already suffered from this issue.
Is there any problem if SITL starts to use 1,500 as the default throttle value?Min/max throttle value
There are user parameters to limit each RC channel such as RC3_MIN and RC3_MAX.
In order to simulate disconnection of RC channel such as “rc 3 900” command, I don’t think we need to limit the minimum of the RC 3 channel. But, users still put any value to RC 3 such as 65000. It causes decreasing altitude of the drone. Can we respect the min/max parameters when users input values on RC channels?Thanks in advance!
Posts: 1
Participants: 1
@OldGazer wrote:
I’ve had this flight controller for quite a while and up until today it has been flawless.
Because of the current pandemic I have been staying home, and the weather has really sucked, so today I decided to check out my 700mm quad.
I connected Pixhawk to my computer and ever thing was fine. At connection time MP threw up a dialog box about upgrading to AC 4.03, so I did. That’s when stuff hit the fan.
The problem is the main LED is dark. I also have a remote led/usb board and that LED is dark too. The USB works just fine. I know the i2c is working because I have an OLED display and the compass working just fine.
So is this a hardware issue, or did 4.03 puke?
So what gives
Posts: 1
Participants: 1
@smartdave wrote:
If I want to make the Baro sensor in my CAN GPS primary, is it as simple as picking the third Baro? In my case I am using a Cube Orange with 2x5611 Baro sensors and then I have a third in my CAN GPS (5611).
GND_EXT_BUS: External baro bus only mentions I2C buss’s and not CAN ones
Also as far as enumeration of Baro sensors. I know it had been talked about, was anything ever coded for it? I would like to see something similar to how Mission planner shows the compass’s and their order.
Posts: 1
Participants: 1
@Ilya_Smolnikov wrote:
I want to use the holybro gps module on my pixhawk 2.4.8 ardupilot.
but ardupilot can’t see ist8310 compass!
I can solder and many times I checked both the wires and the program.
Next I flashed pixhawk with this http://qgroundcontrol.com/.
and found two compasses in the save file !!!
So ardupilot does not support holybro gps?CAL_MAG0_EN 1
CAL_MAG0_ID 396809
CAL_MAG0_ROT 0
CAL_MAG0_XOFF -0.016937162727117538
CAL_MAG0_XSCALE 0.999690413475036621
CAL_MAG0_YOFF -0.023416996002197266
CAL_MAG0_YSCALE 0.980714559555053711
CAL_MAG0_ZOFF -0.086665891110897064
CAL_MAG0_ZSCALE 1.003495216369628906
CAL_MAG1_EN 1
CAL_MAG1_ID 131082
CAL_MAG1_ROT -1
CAL_MAG1_XOFF 0.303862065076828003
CAL_MAG1_XSCALE 0.988722085952758789
CAL_MAG1_YOFF 0.222621470689773560
CAL_MAG1_YSCALE 0.973746120929718018
CAL_MAG1_ZOFF -0.071751624345779419
CAL_MAG1_ZSCALE 1.046743869781494141
CAL_MAG2_ID 0
CAL_MAG2_ROT -1
CAL_MAG3_ID 0
CAL_MAG3_ROT -1
CAL_MAG_PRIME 396809
CAL_MAG_SIDES 63
Posts: 1
Participants: 1
