基于网络化异构冗余执行机构的快响卫星姿态控制方法

With the rapid assembly, low-cost and multi-mode application requirements of operationally responsive satellites, the heterogeneous redundant actuators connected by bus network may encounter the problems of control torque distribution, fault-tolerant control and bus data congestion. In view of the heterogeneous redundant torque distribution problem, this project will adopt ellipsoidal uncertainty sets to describe model uncertainties quantitatively, establish a multi-objective, multi-coupled and multi-constrained moment optimal allocation model, propose a fast global parameter optimization algorithm against complex environments and seek the optimal configuration of this optimization algorithm in convergence speed and robustness.The health management and self-healing fault-tolerant attitude control method based on off-line and on-line data-driven strategy is studied so that the control system has the ability of recovering and maintaining acceptable system performance in case of unknown fault.Change the way of periodically sending control commands, design event-triggered mechanism under the constraint of strongly nonlinear hybrid system and explore new satellite attitude control method based on event-triggered communication mechanism to reduce the bus communication burden.The purpose of this project is to solve the common fundamental problem that the networked hybrid redundant actuators of operationally responsive satellites bring by in attitude control aspect. The research achievements will provide necessary theoretical support to improve the effectiveness of operationally responsive satellites.

针对快速响应卫星的快速组装、低成本、多模式应用的需求，以总线网络连接的、异构冗余的执行机构在实际运行中可能遇到控制力矩分配、容错控制以及总线数据拥堵的问题。针对异构冗余力矩分配问题，采用椭球不确定集对模型的不确定性进行描述，建立多目标、多耦合、多约束的力矩优化分配模型，提出一种面向复杂环境的全域参数快速优化算法，并寻求该优化算法在收敛速度和鲁棒性上的最优配置；研究基于离在线数据驱动的健康管理与姿态自愈合容错控制方法，使控制系统在发生未知故障时具有恢复并保持可接受系统性能的能力；改变周期性发送控制指令方式，在强非线性混杂系统约束下设计事件触发机制，探索基于事件触发通讯机制的卫星姿态控制新方法，降低总线通讯负担。本项目旨在解决快速响应卫星网络化异构冗余执行机构所带来的姿态控制方面共性基础科学问题，项目的研究成果为快速响应卫星应用效能提升提供必要的理论支持。

针对传统的多飞轮能量最优分配策略中可能出现的某个飞轮力矩过大的问题，提出了一种系统能量和输出裕度双目标优化的分配策略；针对混合执行机构，以CMG框架角轴转速及飞轮角加速度、转速有关的能量指标，采用拉格朗日乘子法得到了局部最优的控制力矩分配；提出了一种基于迭代学习观测器的三阶滑模姿态容错控制方法，在存在干扰力矩、系统外部干扰力矩及执行机构故障情况下实现航天器高精度姿态稳定；设计了基于李雅普诺夫方法和严格下凸势函数的姿态控制器，采用自适应滑模控制器解决执行机构饱和与外部干扰，解决了具有多种姿态约束区域的航天器姿态控制问题；设计了一种自适应事件驱动控制方案，并引入径向基函数神经网络来逼近未知非线性项的上界以减少了计算量，降低了控制系统数据的传输速率。项目的研究成果为低成本卫星控制系统性能提升提供必要的理论支持。