动态事件触发下的一阶非线性多智能体系统设定时间一致性

Dynamic event-triggered predefined-time consensus for first-order nonlinear multiagent systems

  • 摘要: 为了保证多智能体系统收敛速度快的同时节约通信资源,针对一阶非线性多智能体系统提出了动态事件触发设定时间一致性算法. 相较于传统的有限时间和固定时间控制算法,设定时间控制算法引入时变函数,可以保证系统在任意预设时间内均可达到一致性. 同时,设定时间控制算法的收敛时间上界不依赖于系统初始状态、拓扑结构和控制参数,可以根据用户需求任意提前设置. 故设定时间控制算法解决了计算冗杂的问题,同时可以保证控制精度. 为了进一步减少触发次数,在现有的静态事件触发控制算法的基础上引入新的动态变量,进而设计出动态事件触发算法. 动态变量随时间改变,可以降低触发次数,从而减少控制器更新频率和执行机构的磨损程度. 鉴于非线性系统广泛存在于实际应用场景中,基于非线性系统设计的控制算法更具有一般性. 最后,通过数值仿真验证动态事件触发设定时间一致性算法的有效性. 综上,本文所提出的动态事件触发设定时间一致性算法可以保证一阶非线性系统在任意设定时间内实现一致性,并且减少触发次数,更可能应用于实际工程.

     

    Abstract: To ensure fast convergence speed and save communication resources, a dynamic event-triggered predefined-time consensus algorithm is proposed for nonlinear multiagent systems in this study. First, the input controller of each agent is developed using a time-varying function to ensure the realization of the predefined-time consensus. Afterward, the measurement error, based on a hyperbolic tangent function, is proposed to exclude the occurrence of Zeno behavior. Finally, a trigger function is established based on the measurement error, and internal dynamic variables are introduced to propose a dynamic event-triggered control protocol, which can reduce triggering times. In view of the widespread presence of nonlinear systems in practical scenarios, the control algorithm based on the developed first-order nonlinear multiagent system is generalized. For the predefined-time control algorithm, a time-varying function is introduced to ensure that the system can achieve consensus at the predefined time. Compared with the conventional finite-time and fixed-time control algorithms, the upper bound of convergence time does not depend on the initial state, topology information, and control parameters but can be arbitrarily set in advance as per the actual needs of users. Thus, the predefined-time control algorithm reduces the computation complexity and ensures control accuracy to a great extent. To further reduce the triggering times, the dynamic event-triggered control algorithm is improved by introducing a novel dynamic variable based on the existing static event-triggered control algorithm. This dynamic variable can be adjusted in real-time with the systems and is always greater than 0. Using the proposed dynamic event-triggered algorithm greatly reduces the triggering times, and Zeno behavior can be eliminated. Therefore, the dynamic event-triggered control algorithm plays an important role in reducing the number of controller updates and actuator wear. Above all, the proposed dynamic event-triggered predefined-time consensus algorithm can not only ensure the convergence speed but also reduce the update frequency of the controller and eventually achieve the consensus in a predefined time without Zeno behavior. Because the control input and measurement error are developed by using the hyperbolic tangent function and the hyperbolic tangent function can solve nondifferential problems caused by discontinuous functions, the nondifferential problems can be excluded. Finally, the effectiveness of the proposed dynamic event-triggered predefined-time consensus algorithm and theoretical analysis results are confirmed using numerical simulations. To sum up, this dynamic event-triggered predefined-time consensus algorithm can ensure the consensus of first-order nonlinear mutiagent systems in arbitrary setting time, and reduce the trigger times, which is more likely to be applied to practical engineering.

     

/

返回文章
返回