Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • 10:47
    dji-dev commented #474
  • 06:28
    dji-dev commented #473
  • 06:28
    dji-dev commented #472
  • Oct 21 13:49
    addriumruss opened #474
  • Oct 21 08:20
    addriumruss opened #473
  • Oct 21 07:50
    addriumruss closed #471
  • Oct 21 07:50
    addriumruss commented #471
  • Oct 21 06:57
    addriumruss commented #472
  • Oct 21 06:49
    dji-dev commented #472
  • Oct 21 06:49
    addriumruss opened #472
  • Oct 11 09:25
    dji-dev commented #471
  • Oct 11 05:38
    addriumruss opened #471
  • Oct 09 11:50
    dji-dev commented #469
  • Oct 09 11:44
    dji-dev commented #470
  • Oct 08 13:20
    liu978079474 commented #469
  • Oct 08 13:14
    liu978079474 commented #469
  • Oct 08 12:27
    hdino opened #470
  • Oct 08 11:14
    dji-dev commented #469
  • Oct 08 03:54
    liu978079474 opened #469
  • Sep 23 20:54

    pcler on 3.9

    (compare)

Aqua
@lanyusea
hello there
amiltonwong
@amiltonwong
Hi, @lanyusea , what will be discussed here?
Aqua
@lanyusea
questions & answers, discussions & replies, suggestions & issues. nothing special, just treat is as a traditional IRC channel ;)
amiltonwong
@amiltonwong
OK, thanks :)
amiltonwong
@amiltonwong
Will DJI release Prop Guards for M100 platform?
Aqua
@lanyusea
maybe......
Andre Nguyen
@andre-nguyen
Hi!
anyone here participating in the International Aerial Robotics Competition?
Aqua
@lanyusea
I was. in last year
I remember there are quit many teams using M100 as their drones this year. The champion and the 2nd runner up in Asia contest.
Andre Nguyen
@andre-nguyen
Cool! I competed in north america these past two years with team Polytechnique Montreal
I heard that this year some teams used the matrice at the Asia arena
But I can't find any videos of what happened
Andre Nguyen
@andre-nguyen
and the top teams didnt submit papers :(
rocketsingh
@robnsngh
@andre-nguyen You can checkout the papers published at iarc at http://www.aerialroboticscompetition.org/symposium2013.php
Almost all the teams have submitted their papers.
Andre Nguyen
@andre-nguyen
naw for 2015 Hong Kong didn't submit one
rocketsingh
@robnsngh
oh snap
Andre Nguyen
@andre-nguyen
They're not the only ones.
Some teams even didn't show up like Madrid and Tsinghua. Oh well...
rocketsingh
@robnsngh
On another note, I think it would be a great idea if DJI could share PID value for the position and altitude controller. Almost everyone will have to tune these controllers and it's just a grunt work for most part.
Andre Nguyen
@andre-nguyen
I've been wondering about that, I did a quick flight test and the m100 flies ridiculously well, does it even use a PID controller?
rocketsingh
@robnsngh
Ofcourse it does. I am talking about pid controllers in case developers want to deploy their own custom rig of lidars and cameras.
Andre Nguyen
@andre-nguyen
Hhhmm I remember that the naza m-lite allowed to modify gains in percentages, but the N1 assistant doesn't allow that. Was thinking maybe the N1 had an adaptive controller or something.
rocketsingh
@robnsngh
@andre-nguyen Interesting, I wasnt aware of that.
rocketsingh
@robnsngh
@lanyusea : I am trying to manipulate the altitude while having a HORI position hold. The controls I want to write is Combination #14 mentioned in Appendix. In combination number 14, the quad does not hold its position for roll and pitch value of 0. It seems a little ambigious from the documentation which specific control_mode_byte allows for position. I tested all control_mode_bytes and only the ones where thrust is not sent as the thrust% seems to be having position control. Having position control while tuning altitude PID makes the process easier. Wasnt sure if this was a common issue so I did not open a new issue and wanted to discuss it out here first
Aqua
@lanyusea
@robnsngh will the 2nd combination suit you?
@robnsngh >In this combination´╝îif all input data is '0', the UAV will brake and hold in a self-balance status at a fixed position.
rocketsingh
@robnsngh
@lanyusea : Since I am using my own height measurement sensor + i am flying indoors, I think the most appt control would be vertical thrust. In theory I think I can hack around my way using verticle velocity command but I would prefer thrust command.
Aqua
@lanyusea
@robnsngh did you install Guidance? cause M100 needs the optical flow sensor on Guidance to hold its position.
rocketsingh
@robnsngh
@lanyusea I am using a 3D lidar :)
for height and position estimation
Aqua
@lanyusea
@robnsngh em... therefore you need your own logic to stabilize the drone. You are flying M100 in the 'A' mode actually.
rocketsingh
@robnsngh
@lanyusea : i understand. For the tuning purpose I am flying M100 outdoors and I was hoping that HORIZ_POS or HORIZ_VEL control byte will be able to hold the position using the GPS (given the health_flag > 3, which is true in my case) . I dont want M100 to move horizontally while I am changing altitude. If I can somehow lock horizontal positions while writing thrust % to the Z -axis, my tuning process would be much easier, is there somehow we can do that ?
Aqua
@lanyusea
@robnsngh I understood. The only option is combination 2, where you can set x/y 0 and turning z with a horizontal position lock. But the problem is what we talked before, no thrust%, only z velocity
rocketsingh
@robnsngh
Thanks ! I assume the Z velocity is computed using onboard barometer (if guidance is unavailable). And indoors, this command might be erroneous since M100's reported Z velocity will change due to changing surroundings. But more or less, it should be still possible to use this way of controlling the altitude since I am using an external sensor to measure and close loop ?
Aqua
@lanyusea
@robnsngh yes possible.