This paper presents master-slave stochastic multi-parameters divergence optimization (SMDO) method for tuning of fractional order proportional-integral-derivative (FOPID) controllers. Bode's ideal control loop (BICL) is used as a reference model that guides the optimization process. Master-slave SMDO performs optimization of FOPID and BICL parameters together. In the first stage called as locking, step responses of FOPID and BICL approximate each other. In the second stage called as drifting, BICL guides tuning of the FOPID towards a desired step response. One of the advantages of presented optimization approach is that both optimization of the reference model (BICL) and controller (FOPID) cooperates and the tuning of FOPID is governed by BICL in the drifting stage. Operation and performance of the proposed algorithm are demonstrated by simulation examples.