This paper presents the optimization of a variable structure controller based on a decentralized fuzzy logic sliding mode (DFLSM) strategy for the control of the pitch and azimuth angles of a TRMS when coupling vertical and horizontal axes. Particle Swarm Optimization (PSO) algorithm is used to tune the DFLSMC parameters to ensure optimal performance in the closed loop, within a given cost function. Finally, simulation results show the effectiveness of the decentralized FLSMC-based PSO algorithm in the stabilization and trajectory tracking tests compared to a non-optimized DFLSMC.