CaΣoS v1.0.0

This is the first version of CaΣoS!

Highlighted Features

• Polynomial data types PD and PS with constant (double) and symbolic coefficients, compatible to CasADi's DM and SX matrix types, respectively.
• Formulate parametrized convex sum-of-squares optimization problems and solve via relaxation to conic programs.
• Formulate parametrized quasiconvex sum-of-squares optimization problems and solve via bisection.
• Formulate parametrized nonlinear and nonconvex sum-of-squares optimization problems and solve via our sequential quadratic method.
• Specify different conic constraints, including second-order cones, positive semidefinite and (scaled) diagonally dominant matrices, and sum-of-squares and (scaled) diagonally dominant sum-of-squares polynomials.
• Define symbolic functions between polynomials and integrate into polynomial expressions.
• Convert polynomials (resulting from a sum-of-squares program) to CasADi functions for further use, e.g., in optimal control.

Dependencies

• This version requires CasADi v3.6.x.
• It is highly recommended to have at least one conic solver installed.

Full Changelog: https://github.com/iFR-ACSO/casos/commits/v1.0.0