Collective behaviors of biological swarms have attracted significant interest in recent years, but much attention and efforts have been focused on constant speed models in which all agents are assumed to move with the same constant speed. One limitation of the constant speed assumption without attraction function is that global convergence is quite difficult or even practically impossible to achieve if the speed is relatively fast. In this paper, we propose an adaptive velocity model in which each agent not only adjusts its moving direction but also adjusts its speed according to the degree of direction consensus among its local neighbors. One important feature of the adaptive velocity model is that the speeds of all agents are adaptively tuned to the same maximum constant speed after a short transient process. The adaptive velocity strategy can greatly enhance the global convergence probability, and thus provides a powerful mechanism for coordinated motion in biological and technological multiagent systems.

• Received 11 June 2006

DOI:https://doi.org/10.1103/PhysRevE.75.021917