Abstract: This work proposes an enhancement to the optimization technique that mimics the flight of a flock of birds, the Particle Swarm Optimization. In the original form of the algorithm, each bird represents a candidate solution and updates its position in the search space from the previous best evaluated positions obtained by itself and by the flock. The proposed innovation is given by the addition of an atmospheric turbulence that affects in an independent, random and sporadic way the flight of each bird of the flock. Optimization tests were performed, using standard real functions, in order to check the new algorithm. It can be concluded that the addition of turbulence is effective, in comparison to the original algorithm, to escape from local minima and to reach better solutions. In addition, the new algorithm is more robust concerning the choice of tuning parameters that balance the influence of the past better positions.