This paper investigates the problem of synchronizing a mobile agent network by means of a velocity adaptation strategy, where each agent is assigned different moving velocities to establish a time-varying network topology, and the velocity of each agent develops adaptively according to the local property between itself and its neighbors. We show that our strategy is effective in enhancing the synchronizability of the mobile agent network, i.e., the region of power density for which the network can achieve synchronization is enlarged as compared to the fast-switching case. In addition, the influence of the controlling parameter on network evolution is studied by assessing the convergence time.