We need to explain why our control model can't be closed loop stable with certain plant parameters (esp. speed)

moorepants / optimal-handling-bicycle-paper

Creative Commons Attribution 4.0 International

0 stars 0 forks source link

Open moorepants opened 7 years ago

moorepants commented 7 years ago

could make root locus plot as a function of two gains: family of curves
if we place the poles for the two first loops to give a neuromuscular peaking for certain vehicle params this isn't a good spot to get closed loop stability after the other two/three loops are closed.

moorepants commented 6 years ago

conservative model (this is a good thing)
at low speeds and for really weird bikes that can be balanced in real life the dynamics are likely dominated by non-linear motion and thus our controller is not capable of controlling (we can't find gains to stabilize the system)
we do not include bicycles that are uncontrollable by our control model in the search space
we need to convince them that our search space is expansive and is applicable to many realizable bicycles

moorepants commented 6 years ago

Ron says that the bike dynamics is the toughest dynamics he's had to control (compared to many aircraft)

moorepants commented 6 years ago

if our model can't compute it it likley has bad handling qualities so we eiimate it from the search space.