Developer

Calling all Tri-Copter pilots

Hi all,

I have finally put a Tri-Copter together in an attempt to get Tri's up to the same level of performance as the more common quad, Y6, X8, Hex and Octo. This is what I have put together:

3691237806?profile=original

3691237821?profile=original

based on these Hobbyking parts:

http://www.hobbyking.com/hobbyking/store/__76263__Hobbyking_8482_Titus_600_Carbon_Fiber_Tri_copter_Frame_With_Integrated_PCB.html

http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=66417

http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=14458

http://www.hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=65156

The tricopter has tuned up reasonably well using autotune and I feel happy with the performance so far. I didn't find the initial setup much more complicated compared to normal quad. I did find the first flight much more nerve racking because of the poor yaw performance during take off. I am interested to see if my confidence in the yaw performance improves now I have a reasonable yaw tune.

I am looking for feedback on what issues Tri-Copter pilots are having with Arducopter. Some of the issues I see are:

1. Dynamics of having the rear motor and propeller CG above the pivot point causes the opposite rotation to the thrust.

2. Roll, pitch, yaw and lift coupling potentially causing issues with auto-tune.

3. The frame type needs a stability patch to prioritize motor output.

4. There is no thrust adjustment on the rear motor based on angle.

5. Yaw auto-tune could be made more efficient by removing any D term or filter frequency tuning and going directly to Rate P tuning.

6. There may be improved motor mixing that does a better job at isolating the 4 control axis.

So let me know what you think I should be looking at that is particularly relevant to Tri-Copters.

You need to be a member of diydrones to add comments!

Join diydrones

Email me when people reply –

Replies

  • Leonard,

    It would be a pleasure to help you test the upcoming tricopter code changes. I own a small tricopter ( quanum trifecta ) and I´m struggling to have authority over yaw without tail wag side effect.

    Autotune on yaw  was unable to help me. It ends with DATA_AUTOTUNE_REACHED_LIMIT ev code which I was unable to figure out why/what it means.

  • Hi Leonard,

    I haven't seen any activity on this effort in a while.  Is there anything we can do with our tricopters to help with your efforts? 

    On the tail wag issue I've seen two types on my tricopter - a slow wag which I would very much like to eliminate and a ringing type of oscillation that I've sometimes seen.   I wasn't which type of wag is affected by center of gravity on the motor with the servo control.

  • Leonardt, could you please verify how is mot_sv_yaw_trim is used for the tail control? Is it used at all? In the previous versions I didn't have to adjust ch7 mid parameter. Please advise, thank you! 

    P.s. in cleanflight one needs to manually dial yaw trim on the first flight, would it be possible to fully disable yaw stabilization while in autotrim this would allow to trim yaw very precisely. Also, there is a cleanflight fork which couples yaw and roll would you think it makes sence to account for yaw affect during roll and vice versa? 

    • Developer

      Hi Artem,

      Sorry, I couldn't find that variable anywhere. I may have missed something. I will have to look into that and get back to you.

      I will have a chat to Randy about some of your other suggestions and but I will keep everything in mind for the next release. However, my first priority is to ensure that tri copters have all the basic functionality they need. So at the moment I will be focussing on the stability patch for tri's, basic controller design for yaw, and servo setup (that will include the update frequency). Polishing stuff like pinout will need to wait unless I see a quick and easy way to do it.

      My understanding is that cleanflight uses a rate only controller on the yaw so it has different issues. Our I term takes care of this offset very quickly so it may not be needed at all. I will look at this as I work through the issues.

      As for Yaw to roll coupling this is something I may be able to add for Tri's I will look at this as I work forward.

       

  • Leonardt, please consider the suggestions here, some of them are very simple and can be included in 3.3.1 if not 3.3 final.

    1. please move servo to channel 2 output. here is why:

             a. It will make setup much cleaner and can improve vibration passing through wires by using only 2 "3pin servo connectors"  like so _|  (power for servo should be harnessed directly off the BEC so vcc/gnd pins not needed for servos)

             b. besides the aesthetics/convenience such setup will work with the inevitably coming 36x36 versions of pixhawk which will have room only for 6 output pins, this is for the racer quads/tries, mini connectors are horrible on those as people are crashing, ESCs go out all the time and it is a huge PITA to change the whole harness (as with mini connectors) 

              c. one might ask - why not rearrange the whole output to  mot1 mot2 mot3 servo4? well this is logical, but can potentially be safety issue as discussed in the RobL's tricopter thread 

    2. Please allow to change servo output frequency as right not it is hard-coded to 50mhz which is OK for regular servos, but regular servos suck on mini tricopters (~280size) use of better servos (helicopter) will greatly improve yaw performance, however, their optimal pwm frequency varies from 50mhz all the way to 560mhz, I believe necessary for this code exists in heli version of AC. 

    3. Please make default rate_yaw_filt_hz parameter 20mhz and I max 200.  these two are MUST on 3.3 as otherwise yaw is barely controllable. 

    Thank you! 

  • Yaw Wag can be down to the servo frequency and speed ., but  to minimize the load on the servo I keep the CG forward so the tail motor will only stabilize the copter the most of the lift are from the two mains

  • Moderator

    First quick flight in gusty conditions with v3.3rc10 directly after setup with a AUAV-X1 flight controller on a small carbon tricopter. No problems encountered except for a persistent compass variance error which required COMPASS_ORIENT2 to be set to 8 (Roll180), the same as AHRS_ORIENTATION on these boards. Also had a couple of bad AHRS errors initially but they went away. The log file is attached.

    Auto Analysis with Mission Planner didn't work for some reason with this error: Could not add reference to assembly mtrand.dll

    2015-09-02 08-43-13.bin

  • Here's one for you, guys. Have you found autotune near impossible with a tri due to drift during alt-hold? I used to think it was only related to breeze but I tried tonight in near absolute calm, and it continually drifted away as if it had a persistent moderate tailwind. It's enough that I don't think autotune would even be possible without hundreds of acres of empty space (quite a premium anywhere near a metro area). I have tried letting it tune then pulling it back any time it nears an obstacle, but I have found the start/stop tune like that leads to poor results.

    Has anyone else seen this issue, and is it something I can account for somehow in order to allow for an autotune?

    • Developer

      Hi Patrick,

      Sounds like you need to do an in-flight level. There are a couple of methods described on the gui.

      I have a plan that will automatically stop the drift if you enter Autotune from loiter. But I have not done that yet.

This reply was deleted.

Activity

DIY Robocars via Twitter
RT @TinkerGen_: "The Tinkergen MARK ($199) is my new favorite starter robocar. It’s got everything — computer vision, deep learning, sensor…
20 hours ago
DIY Robocars via Twitter
20 hours ago
DIY Robocars via Twitter
RT @roboton_io: Join our FREE Sumo Competition 🤖🏆 👉 https://roboton.io/ranking/vsc2020 #sumo #robot #edtech #competition #games4ed https://t.co/WOx…
Nov 16
DIY Drones via Twitter
First impressions of Tinkergen MARK robocar https://ift.tt/36IeZHc
Nov 16
DIY Robocars via Twitter
Our review of the @TinkerGen_ MARK robocar, which is the best on the market right now https://diyrobocars.com/2020/11/15/first-impressions-of-tinkergen-mark-robocar/ https://t.co/ENIlU5SfZ2
Nov 15
DIY Robocars via Twitter
RT @Ingmar_Stapel: I have now explained the OpenBot project in great detail on my blog with 12 articles step by step. I hope you enjoy read…
Nov 15
DIY Robocars via Twitter
RT @DAVGtech: This is a must attend. Click the link, follow link to read the story, sign up. #chaos2020 #digitalconnection #digitalworld ht…
Nov 15
DIY Robocars via Twitter
RT @a1k0n: Got a new chassis for outdoor races (hobbyking Quantum Vandal) but I totally didn't expect that it might cause problems for my g…
Nov 11
DIY Drones via Twitter
First impressions of the Intel OpenBot https://ift.tt/36qkVV4
Nov 10
DIY Robocars via Twitter
Nov 9
DIY Robocars via Twitter
Excellent use of cardboard instead of 3D printing! https://twitter.com/Ingmar_Stapel/status/1324960595318333441
Nov 7
DIY Robocars via Twitter
RT @chr1sa: We've got a record 50 teams competing in this month's @DIYRobocars @donkey_car virtual AI car race. Starting today at 10:00am…
Nov 7
DIY Robocars via Twitter
Nov 6
DIY Robocars via Twitter
RT @a1k0n: Car's view, using a fisheye camera. The ceiling light tracking algorithm gave me some ideas to improve ConeSLAM, and having grou…
Nov 5
DIY Robocars via Twitter
RT @a1k0n: To get ground truth I measured the rug, found the pixel coordinates of its corners, calibrated my phone camera with my standard…
Nov 5
DIY Robocars via Twitter
RT @a1k0n: @DIYRobocars is back in December, but outside. Time to reinvestigate ConeSLAM! I rigged up a quick and dirty ground-truth captur…
Nov 5
More…