摘要
为解决Flocking模型中势函数没有考虑角度信息的问题,建立了基于加权准邻域势函数以及多虚拟领航的Flocking运动模型。首先,考虑运动的方向性,不同角度的邻居对智能体的影响不同,根据邻域角度信息设计基于加权势函数的控制器。然后,对单虚拟领航者的群智能体运动和两组群智能体受各自领航者引导混杂后又分离的运动过程进行了仿真。最后,通过误差分析证明所建立的Flocking运动模型符合群智能体的运动规律。
The improvement of the flocking model in which the potential function does not take the angle information into account leads to the establishment of the flocking movement model based on the potential function with weighted quasi neighbors and multi-virtual leaders. First, due to the direction of movement, neighbors of different angles have different effect on the agent and the controller is de- signed based on the weighted potential function of angle. Then, the movement of flocking with one virtual leader is simulated. The immingling and segregating processes of two flocking guided by each virtual leader are simulated too. Finally, the error analysis is made and the result proves that the flocking model based on the weighted quasi neighbor can meet the movement rule of multi-agent.
出处
《海军工程大学学报》
CAS
北大核心
2013年第5期88-94,共7页
Journal of Naval University of Engineering
