The paper deals with a biologically inspired model of self-propelled particles introduced by Vicsek. To solve the problem of low convergence efficiency in this model, an improved model based on distance weight is proposed in this paper. Particularly, distance weight function is designed in the form of polynomial function which is a monotone increasing function of distance. Moreover, a new index to evaluate the convergence efficiency called Vicsek algebraic connectivity is promoted. Finally, comprehensive comparative studies of the convergence properties among the improved model, original Vicsek model, and Degree model are investigated in the simulation part. The simulation results show that our modified model is better than other two models in convergence probability and consensus time. Our results may enlighten other researchers in revealing the mechanism of collective motion.