文献类型：期刊文献

中文题名：垂直起降无人机量化通信下分散式带记忆控制

英文题名：Memory-Based Decentralized Control for PVTOL Drones Under Quantitative Communication

作者：张辰[1];许璟[1]

机构：[1]华东理工大学信息科学与工程学院能源化工过程智能制造教育部重点实验室,上海200000

年份：2026

卷号：33

期号：2

起止页码：48

中文期刊名：电光与控制

外文期刊名：Electronics Optics & Control

收录：;北大核心:【北大核心2023】；

基金：国家自然科学基金(62333006,62173141);上海市自然科学基金(22ZR1417900)。

语种：中文

中文关键词：网络化控制系统;滑模控制;线性矩阵不等式;人工时滞

外文关键词：networked control system;sliding mode control;linear matrix inequality;artificial time delay

摘要：垂直起降(PVTOL)无人机集群存在通信带宽受限、无人机间互相干扰等问题。网络化控制系统中,带宽限制和系统状态量化问题更为明显,尤其是在复杂的分散式系统中。通过量化的方法可以有效减少网络带宽需求,但是量化误差增加了系统稳定性分析的复杂性。带记忆控制的方法利用人工时滞和泰勒展开减少对系统状态的依赖,但参数设计复杂。针对上述问题,使用一种新的带记忆滑模反馈控制器设计方法构建线性矩阵不等式框架,并使用Lyapunov-K rasovskii泛函证明了系统的稳定性。此外,采用优化算法优化控制器参数,在尽可能少占用带宽的情况下,更快速、更精确地使系统收敛到最小的稳定区域。在PVTOL无人机集群中,通过对比实验验证了所提出的控制方法的有效性。
Planar Vertical Take-Off and Landing(PVTOL)drone swarms face challenges such as limited communication bandwidth and mutual interference.In networked control systems,bandwidth constraints and system state quantization are particularly significant,especially in complex decentralized systems.Quantitative methods can effectively reduce the demand for network bandwidth,but the presence of quantization errors increases the complexity of system stability analysis.Memory-based control method uses artificial delays and Taylor expansion for state estimation,reducing reliance on system states,but the design of parameters is complicated.In order to solve above problems,a novel design method of memory-based sliding mode feedback controller is applied to construct a linear matrix inequality framework,and Lyapunov-Krasovskii function is used to prove the global system stability.In addition,an optimization algorithm is developed to optimize controller parameters,enabling the system to converge rapidly and precisely to the smallest stable region while minimizing bandwidth usage.The effectiveness of the proposed control method is validated through comparative experiments in PVTOL drone swarms.