svigerske
commented
4 years ago Comment by @apotschka created at 2017-10-04 11:05:08
Please refer to the following publication, which we kindly request everybody to cite for their results obtained with qpOASES (as stated on https://projects.coin-or.org/qpOASES):
Ferreau, Kirches, Potschka, Bock, Diehl, "qpOASES: a parametric active-set algorithm for quadratic programming", Math. Prog. Comp. (2014) 6:327–363, DOI 10.1007/s12532-014-0071-1
For your convenience, it is available here:
http://mpc.zib.de/index.php/MPC/article/download/151/73
Visualization is helpful only for very small problems (2D or 3D). Hence, there is no built-in method available in qpOASES. I suggest you set up your own sequence of QPs along a given affine-linear homotopy: Take tau from an equidistant grid of the interval [0, 1] and solve a sequence of QPs with constant Hessian and constraint matrices, but with objective gradients and variable/constraint bounds of the form g = g0 + tau (g1 - g0). You can then visualize the "solution curve" for tau going from 0 to 1, which is essentially what qpOASES does.
Issue created by migration from Trac.
Original creator: flohgroh
Original creation time: 2017-10-04 08:31:51
Assignee: ferreau
Version: 3.2.1
Dear Sir or Madam,
I'm working on my master thesis and want to use qpOASES to optimize the control of a pump system. I read through the Manual and some other sources from qpOASES but wasnt able to find a description about the functionally/mathod of working of the algorithm. Most times I just found how I can implement the c code and how to use it but i want to visualize how the algorithm finds the optimum solutions.
Do you have any descriptions about the process of the algoritm which you can send to me?