仅用位置信息的多智能体自适应蜂拥控制

Adaptive Adjusted Flocking of Multi-Agents Based on Positions Only

复杂系统的研究中,提出了自适应调节的多智能体蜂拥控制算法。算法首先通过改进人工势函数来保证网络的连通性。再通过构造分布式观测器,用多智能体之间的位置信息来估计多智能体之间的速度信息;同时对观测器权重参数进行自适应调节来降低蜂拥控制模型的参数不确定性的影响。并且从理论上对算法进行了分析,然后给出计算机仿真,证明了算法的有效性。仿真结果表明,上述算法能够使多智能体经过一定时间的调整达到速度一致并进入稳态,且运动过程中保证网络连通性。

On the scope of the study of complex systems, an algorithm for adaptive adjusted flocking of multi - a- gents was proposed in this paper. The algorithm was designed as follows. Firstly, the paper constructed the artificial potential functions to preserve connectivity. Secondly, distributed observers were used to estimate the speed between agents by the positions. At last, through adjusting the weights of observer, the affect of parameter uncertainties on the flocking control model was reduced. The algorithm was theoretically analyzed and simulation was used to prove effec- tiveness of the proposed algorithm. The simulation results show that after a certain period of adjustment, the algorithm makes the multi- agents reach the same velocity and come into steady state and guarantees the network connectivity while moving.