IEEE Trans Cybern. 2016 Jul;46(7):1655-66. doi: 10.1109/TCYB.2015.2453167. Epub 2015 Aug 25.
The paper studies the distributed optimization problem for a class of nonlinear multiagent systems in the presence of external disturbances. To solve the problem, we need to achieve the optimal multiagent consensus based on local cost function information and neighboring information and meanwhile to reject local disturbance signals modeled by an exogenous system. With convex analysis and the internal model approach, we propose a distributed optimization controller for heterogeneous and nonlinear agents in the form of continuous-time minimum-phase systems with unity relative degree. We prove that the proposed design can solve the exact optimization problem with rejecting disturbances.
本文研究了存在外部干扰时一类非线性多智能体系统的分布式优化问题。为了解决这个问题，我们需要基于局部代价函数信息和邻居信息实现最优多智能体一致性，同时拒绝由外部系统建模的局部干扰信号。利用凸分析和内模方法，我们提出了一种具有连续时间最小相位系统和单位相对度的异构非线性智能体的分布式优化控制器设计。我们证明了所提出的设计可以解决具有干扰抑制的精确优化问题。
