In this article, the problem of appropriate active load sharing, frequency regulation, and adequate damping in an autonomous voltage source converter based microgrid, which has a decentralized controller, is investigated. In decentralized autonomous voltage source converter based microgrids, proper load sharing needs a high gain frequency droop. However, this high gain has a negative impact on small-signal stability and frequency regulation. In order to eliminate this drawback, an auxiliary loop around the frequency droop method is proposed to increase oscillation damping and to decrease frequency deviation. Tuning of this auxiliary loop for each micro-source is formulated as an off-line fuzzy multi-objective optimization problem, which is solved using an evolutionary algorithm. The simulation results based on frequency-domain modeling and eigenvalue analysis show the oscillation damping improvement, frequency deviation reduction, and active power-sharing accuracy of the microgrid in the proposed methodology for wide range of operation conditions.