Automatic Synthesis of a Saturating State Space Controller Based on Convex Optimization for Task Space Controlled Industrial Robots

Lorenz Halt , Philipp Tenbrock , Frank Nägele and Andreas Pott

Software contribution to the submitted research paper

This repository holds all necessary data, including the identified model in form of a npy-file (numpy pickle) and the convex optimisation problem itself.

The jupyter notebook loads all data and executes an optimisation. To run it on your own computer please make sure to resolve all dependencies (python 2.7):

• numpy
• cvxpy (incl. solver: CVXOPT, MOSEK(1), SCS)
• matplotlib

(1) A licence is necessary for MOSEK. However, any suitable SDP solver can be included.