具有预设性能的不确定非线性多智能体系统时变/不连续反馈控制设计

Multi-agent systems are not only a way to understand the complex behavior of bees, birds and fish in nature, but also have important applications in robots, sensor networks, smart grids and so on. Due to the limitation of physical conditions and control costs, it is usually necessary to ensure not only the steady-state performances of the practical systems, such as stability and tracking, but also the transient performances of the systems, such as convergence rate and maximum overshoot. For multi-agent systems with serious nonlinearities and uncertainties, this project intends to study the existence and constructive design methods of time-varying or discontinuous feedback control for the systems with stronger feedback ability and better transient and steady-state performances, which cannot be solved by the existing control schemes. The feedback compensation mechanisms of nonlinearities and uncertainties are established by combining backstepping, time-varying high gain, Nussbaum gain, switching and so on. This project also develops time-varying or discontinuous feedback control strategies with prescribe performance and higher communication efficiency. Then the feasibility of the developed feedback control strategies is verified in manipulators and mobile robots. Finally, the developed time-varying or discontinuous feedback control strategies will make up for the shortcomings of the existing methods for uncertain nonlinear multi-agent systems, and provide new ideas and sights for solving scientific problems which can not be solved by the existing control strategies.

多智能体系统不仅是理解自然界蜂群、鸟群、鱼群等复杂群集行为的一个途径，而且在机器人、传感器网络、智能电网等领域具有重要的应用价值。由于物理条件限制、控制成本有限等，通常不仅要保证实际系统的稳定、跟踪等稳态性能，还要保证系统的收敛速度、最大超调等暂态性能。本项目针对带有严重非线性、不确定性的多智能体系统，在现有反馈控制设计框架难以实施的情况下，研究反馈能力更强、暂稳态性能更优的时变/不连续反馈控制策略的可行性和构造性设计方法；综合运用反推、时变高增益、Nussbaum增益、切换等方法建立非线性、不确定性的反馈补偿机制；发展具有预设性能、提升网络通信效率的时变/不连续反馈控制策略；验证发展的反馈控制策略在机械手、移动机器人中的可行性，建立满足实际需求的轨迹跟踪控制算法。本项目所发展的时变/不连续反馈控制策略将弥补现有多智能体系统研究中存在的不足，为现有控制策略不能解决的问题提供新思路、新途径。

多智能体系统不仅是理解自然界鸟群、蜂群、鱼群等复杂群集行为的一个途径，而且在无人系统、传感器网络、智能电网等领域具有重要的应用价值。本项目针对带有不确定性和外部干扰的非线性多智能体系统，利用时变反馈控制方法补偿了系统中的不确定性，实现了系统的有限时间一致性；针对带有未知干扰的非线性系统，利用扰动观测器方法和自适应方法实现了系统的编队控制；针对带有未知干扰的多智能体系统，利用输入到状态稳定和李雅普诺夫稳定性定理，不仅证明了所提控制算法的收敛性和稳定性，抑制了系统的外部干扰，还保证了系统的误差状态收敛到预设的性能函数内，保证系统具有良好的暂态行为；将上述发展的方法，应用到无人系统的编队控制和生物种群的稳定性中。本项目发展的时变/不连续反馈控制策略将弥补现有多智能体系统研究中存在的不足，为现有控制策略不能解决的问题提供新思路。