Next week (tuesday feb 4, 16:00) our PhD student Ilhan Polat will defend his thesis on the use of Robust Control Theory to have a better control system for Haptic devices, such as master/slave medical robotics.
Here is the teaser video of his work!
To explain a little more about the subject, I use some of the material Ilhan showed me in his rehearsal of his PhD 10 min talk. First, the key idea of feedback is often seen explicitly in nature, like the well-know chicken stabilization video:
In industrial robotics we can use similar things, watch this video, especially from 35 seconds onwards:
Now, talking about feedback for haptics, the issue which makes it more difficult, is that on the one hand the person using the master device is to a certain extend an unknown dynamic system, and the environment on the slave side can change, even rapidly. This is called the robustness issue, and is demonstrated very nicely with this video of a self destructing washing machine (start at 1:06 min):
Clearly, the washing machine is not designed to account for throwing in bricks. And that is also sensible, because if it would, the machine would be too costly, and since throwing bricks into washing machines is outside the normal use of it, the machine would be called designed ‘conservatively’. This is similar to our field of control engineering where we want to design for robustness and the challenge is to design robust controllers which are not conservative. This then relates to the question how to model the uncertainty in our systems to be controlled, like the human operator and the environment in haptic systems. In the work of Ilhan, he made use of modern robust control theory to solve this problem for haptics, and his video above shows his results!