2
$\begingroup$

I am planning a tank like robot for hobby purpose. I have control engineering background, however I never applied on robotics. I would like to test different control theory, namely MPC. I saw a lot of publications regarding the kinematics and inverse kinematics of such a robot, however I am wondering if somebody can point out regarding the dynamics modelling of such a system,taking into account the forces, mass etc?

$\endgroup$
  • 3
    $\begingroup$ you might consider asking two different questions. first is related to deriving the mathematical model (dynamics of the tank) and the second is related to sensors in a tank-like-robot and is a different topic. $\endgroup$ – Gürkan Çetin May 8 '15 at 19:39
  • 1
    $\begingroup$ As @GürkanÇetin said I would suggest splitting your question to make it easier to answer, as for the positioning the accuracy depends on what you need, do you need 10m accuracy? do you need centimeter accuracy? what is your target application for this robot? $\endgroup$ – Mark Omo May 10 '15 at 15:26
  • $\begingroup$ I edited the question. I found the answer for the second one, hence I kept the first only. $\endgroup$ – OHLÁLÁ May 10 '15 at 15:27
  • $\begingroup$ The question is currenly asking for parameter estimation, and not system identification. I suppose, for hobby purposes, parameter estimation might be a bit overkill. if you do not wish to have a perfect performing, speed holding tank. Tuning the control loops with tests might be more efficient. (resultant tank hardware and software / time and effort committed) I have not come across a good article or web source, so I started drafting something, but it will take too long and I've got no background on tanks. previous PID tuning discussions might be worth checking. $\endgroup$ – Gürkan Çetin May 10 '15 at 19:17
  • $\begingroup$ Can you explain a bit more what problem you are trying to solve? In other words, what are you trying to accomplish by modelling the system? $\endgroup$ – Ian May 15 '15 at 13:41
2
$\begingroup$

For building a dynamic model from scratch of any differential drive mobile robot (i.e., a tank), the best resource I've found so far is a paper by Dhaouadi and Hatab (PDF link) titled "Dynamic Modelling of Differential-Drive Mobile Robots using Lagrange and Newton-Euler Methodologies: A Unified Framework". It even includes some discussion of how to do the dynamic modeling of the actuators.

| improve this answer | |
$\endgroup$

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.