A. "Data-Driven Superstabilizing Control of Error-in-Variables Discrete-Time Linear Systems"

B. "Data-driven stabilization and robust control of linear systems corrupted by error in variables“

Install YALMIP and MOSEK before running the code

Algorithm 1 in A, test_Full_SS.m

full method for superstability, only works for 1st order system due to high complexity

Algorithm 2 in A, test_Dual_SS.m

dual method for superstability, greatly reduce the complexity, works for 2nd order system

section 7 in B, test_Dual_SS_all_noise.m

dual method considering all type of noises (dx, du, w), this covers test_Dual_SS.m

Algorithm 5,6,7 in B, test_Dual_QS.m

dual method for quadratic stability, worst case H2 works while Hinf currently has bug

Algorithm 5,6,7 + section 7 in B, test_Dual_QS_all_noise.m

dual method considering all type of noises (dx, du, w), this covers test_Dual_QS.m

complexity.m computes the number of constraints and variables

Dual_SS_manual.m manually defines the multiplication in the coefficients, it is much faster than Dual_SS.m while currently only supports SS

Contact Tianyu (dai.ti@northeastern.edu) for any questions regarding the code