dthierry / cappresse

The Control and Adaptation with Pre-Dictive Sensitivity Enhancements

Non-linear model control with python

This is a work in progress to create a framework to perform NLP-based control with Pyomo

Copyright & License

Copyright (c) 2017, David Thierry. MIT License.

cappresse's Issues

Pyomo 5.7 compatibility

All broken.

[py.dae] TODO

• Create reasonable unittest for all the framework
• Implement the .clone() strategy throughout the whole framework
• Deal with the steady-state problem. fetch dvars, dvars==0, delete collocation equations, delete initial conditions
• Dummy param vs fixed var for the input

ngc at the bfb

It should not have the Ax term.

init_step_mhe is not efficient

This needs to be fixed
https://github.com/dthierry/nmpc_mhe_q/blob/fe0as_patch/nmpc_mhe/dync/MHEGenv2.py

we need k_aug to use mumps5 and mc19

Non-licensed stuff; so we can distribute it easily.

Installation failure in windows

The aux folders need to be renamed.

mistake in snippet for nmpc (find_target_ss)

    for i in self.ref_state.keys():
        v = getattr(self.SteadyRef2, i[0])
        val = value((v[(0, 1) + vkey]))
        **vkey = i[1]**
        ofexp += weights[i] * (v[(0, 1) + vkey] - self.ref_state[i])**2

shift_mhe method has a potential bug

To be fixed:
v[k].set_value(v[(k[0] + 1,) + k[1:]])

load_iguess_dyndyn is not implemented properly

The behavior is not the intended one.
One to one mapping for the collocation points was expected but it is not the case.