多智能体网络的一致性分析和分布式控制

This project aims to investigate the consensus analysis and distributed control for positive multi-agent systems (PMASs). Multi-agent systems are a class of networked systems consisting of a group of autonomous agents, which interact with each other to achieve cooperative dynamics over a communication network. PMASs are MASs whose state variables are non-negative. As the development of computing, communication and control technologies, PMASs have been frequently encountered in various fields, such as science and sociology, and have recently drawn much attention of researchers around the world. By combining the theories of both positive systems and MASs, this project will study the modeling of MASs under positivity constraints, analyze the effects of positivity, nonlinearity and communication delays on the network’s consensus, develop various conditions for PMASs to achieve consensus, design distributed controllers for PMASs such that the controller topology and the controller gain can be obtained simultaneously while the positivity is preserved. The outcome of this project will not only enhance the research scope for PMASs, but also enrich the application of positive systems theory to MASs.

本项目研究正多智能体网络的一致性分析和分布式控制问题。多智能体网络是由多个具有独立自主能力的动态个体通过相互通信、彼此协作共同完成复杂作业的网络化系统。正多智能体网络是指系统状态为非负的智能体网络。随着计算机、通信和控制等技术的发展，正多智能体网络的应用已经深入到自然科学和社会科学等各个领域，成为近年来国际前沿的研究课题之一。本项目将结合正系统理论和多智能体系统理论，深入研究具有正性约束的多智能体系统的建模，分析正性、非线性和通信延迟对系统一致性的影响，给出正多智能体系统达到一致性的条件，在保持系统正性的前提下，设计多智能体系统的分布式控制器以同时得到控制器的拓扑结构和参数。本项目的实施将不仅推动正多智能体系统的研究领域，也将进一步丰富和完善正系统理论在多智能体网络中的应用，具有重要的理论意义和实际价值。

本项目研究正多智能体系统的一致性分析和控制问题。多智能体网络是由多个具有独立自主能力的动态个体通过相互通信、彼此协作共同完成复杂作业的网络化系统。正多智能体系统是指系统状态为非负的智能体系统。随着计算机、通信和控制等技术的发展，正多智能体网络的应用已经深入到自然科学和社会科学等各个领域，成为近年来国际前沿的研究课题之一。本项目结合了多智能体系统和正系统理论，研究了正性、非线性和通信延迟对系统一致性的影响，给出了智能体系统达到一致性的条件，设计了智能体系统的分布式控制器。另外，研究了基于多项式逼近的非线性系统控制和一类具有未知延迟的非线性系统的自适应跟踪控制。系统学习了在l1性能指标下，区间不确定性的正系统的控制器设计问题和非线性系统的观测器设计问题。期望本项目的实施能推动多智能体系统的研究领域，并进一步丰富和完善正系统理论在智能系统中的应用。