I am attempting to model a robot and develop a PID based off of the paper at the bottom of this post before developing a physical model. However, the research paper appears to give parameters for wheel radius, a geared motor inertia and a tachometer constant but these are not included in the overall transfer function stated in the paper - unless these would be factored in by the parameters for equivalent damping / inertia?

Were I to include these parameters so to achieve a more realistic model how would that be done? The transfer function relates the speed of the robot to the voltage applied across the DC motor - so I presume this multiplied by the wheel radius can give the velocity of the wheels.

Any help/advice greatly appreciated.

Edit: Section of paper below


  • \$\begingroup\$ Do you think, that we all have a subscription to IEEE? \$\endgroup\$ – Marko Buršič Feb 26 at 18:08
  • \$\begingroup\$ This question is likely to be answered with opinions rather than facts and citations. It should be updated so it will lead to fact-based answers. \$\endgroup\$ – Andy aka Feb 26 at 18:32
  • \$\begingroup\$ Please quote relevant section(s) of the linked paper for users who don't have access to it. \$\endgroup\$ – Null Feb 26 at 19:04
  • \$\begingroup\$ Edited so to include a drive link of the paper vs. IEEE link - thank you \$\endgroup\$ – Brandon Kellett Feb 26 at 19:26
  • \$\begingroup\$ Um, are you now publishing a copyright IEEE paper? \$\endgroup\$ – Transistor Feb 26 at 19:36

enter image description here This is the block diagram of DC motor, similar you should have in your article:

And this is the transfer function of it:

$$ \dfrac{\Omega(s)}{u_q(s)} =\dfrac{{k_\Phi}}{s^2L_qJ + s{R_qJ+L_qF}+{R_qF+k_\Phi^2}} $$

Replacing the upper \$k_{\Phi}\$ in the diagram with \$k_T\$ and the lower one with \$k_v\$ we get almost the same transfer function as your blurry one, except B=F, :

$$ \dfrac{\Omega(s)}{u_q(s)} =\dfrac{{k_T}}{s^2L_qJ + s{R_qJ+L_qF}+{R_qF+k_v\cdot k_T}} $$

You have to place the inertia and friction parameter in the feedforward path of your position controller \$T=J \alpha + F \Omega\$. The approx. amount of torque is known "a priori" with these two parameters known.

Where to place Radius-R(not resistance)? \$l=\Theta R\$; \$\Theta =\dfrac{\Omega}{s}\$


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.