A Local Consistency Algorithm to Shorten the Convergence Time and Improve the Robustness of Self-propelled Swarms

This paper explores an update and iteration rule which enables all the individuals of the particle swarm to reach consistency, via understanding of collective behaviors of biological swarms. This rule takes the local consistency of individuals in the neighborhood into consideration and measures the angle update decision on each individual at the next moment with the expression containing local consistency as the weight. This algorithm is to replace the update strategy of the original Vicsek model. The convergence of time and efficiency between the modified model and the original Vicsek model are contrasted. The use of local consistency to shorten the convergence time is consistent with the phenomenon that individuals in nature make respond opportunely in the neighborhood in order to adapt to migration. Furthermore, the application of noise analysis certifies the stability of the modified model. The simulation results demonstrate that the introduction of local consistency parameter significantly improves the convergence time and efficiency and has good robustness in noisy environment.