1. Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.
    Dismiss Notice

Control system modelling & simulation from dynamic equations (quadrotor)

Discussion in 'Robotics & Mechatronics' started by mikeOJ, Aug 5, 2014.

  1. steveB

    steveB Well-Known Member Most Helpful Member

    Joined:
    Jan 16, 2009
    Messages:
    1,297
    Likes:
    629
    Here you have to decide whether to use the torque/thrust as the inputs to the model, or the motor velocities. The problem with using the velocities is that the equations are then nonlinear (more nonlinear than with torque/thrust) and you would have to linearize the inputs. It is probably better to let the torques/thrust be the inputs to the system, and then use the nonlinear equations to solve for and generate the required velocities. At least that is better assuming you can work out the equations in a nice way that is implementable. If you can't you may need to linearize the velocity inputs to have a usable design model.

    As I said above, I have not tried to work any of this out to see which approach is best, but once you dive in, one method will be better than the other, and you will know which path to take.
     
  2. steveB

    steveB Well-Known Member Most Helpful Member

    Joined:
    Jan 16, 2009
    Messages:
    1,297
    Likes:
    629
    Another comment I can make is that you could allow the square of the velocities be the inputs. This way, the speed squared is like a linear variable in the model.
     
  3. steveB

    steveB Well-Known Member Most Helpful Member

    Joined:
    Jan 16, 2009
    Messages:
    1,297
    Likes:
    629
    Another comment, if you don't mind me brain storming, is as follows.

    If you do adopt the idea in post #22 above, and use the velocity squared as the input to the model, then you can also set up differential-mode, common-mode variable definitions as follows.

    u13p = (ω1^2) + (ω3^2)
    u13m = (ω1^2) - (ω3^2)
    u24p = (ω2^2) + (ω4^2)
    u24m = (ω2^2) - (ω4^2)

    This results in the following

    Tψ is: k ( u13p - u24p )
    Tθ is: L k u24m
    and Tβ is: -L k u13m

    Then the thrust will be something proportional to the sum of all speeds squared such that: u=c ( u13p + u24p ) where c is a constant.
     
    Last edited: Aug 12, 2014

Share This Page