多航天器有限时间姿态协同跟踪与合围控制研究

The cooperative attitude control is the fundamental requirement to achieve spacecraft formation flying (SFF). In the deep space environment, the dynamic model of spacecraft is easily affected by the model uncertainties, external disturbances and actuator faults. In addition, the unreliable communications may occur among spacecraft. The aforementioned factors all have a negative impact on the attitude control accuracy and coordination capability, and even lead to the failure of SFF. Considering these factors, the project aims to develop continuous finite time cooperative attitude control methods for multiple spacecraft under single-way directed communication interactions. Firstly, under single-way directed graph with one leader, an adaptive finite attitude control method is proposed for SFF in the presence of internal uncertainties and external disturbances. The proposed controller can not only avoid the algebraic loop problem, but also reduce the control chattering effectively. Secondly, the cooperative attitude fault-tolerant control problem is further addressed for SFF subject to unreliable communications and actuation failures among spacecraft simultaneously. Moreover, sufficient conditions that guarantee finite-time attitude tracking performance under complicated communication conditions are established. Then, a unified continuous finite-time control framework that suits for both multiple leaders and followers is proposed. In this proposed control framework, the attitude formation adjustments for the leaders can be achieved, and meanwhile, the attitude containment control for the followers can be also realized. Finally, the theoretical results will be validated by an attitude simulation platform of multiple spacecraft. The implementation of this project will provide theoretical and technical supports for engineering practices of SFF.

姿态协同控制是实现多航天器编队飞行的基础。在深空环境下，航天器内部容易受到模型不确定性、外界扰动及执行器故障等因素影响，且航天器之间会出现不可靠的通信交互。这些因素都会影响航天器的姿态控制精度和稳定性，导致编队飞行的失败。针对这些因素，本项目旨在提出单向网络下连续有限时间姿态协同控制方法。首先，在单领航者-多跟随者网络结构下，综合考虑多种不确定性对航天器的影响，提出连续自适应有限时间姿态控制器，不仅避免控制器出现代数环问题，而且有效抑制控制抖振现象；其次，研究执行器故障和不可靠通信下的姿态协同容错控制问题，建立航天器在通信约束下实现有限时间姿态跟踪的判定条件；再次，研究多领航者和多跟随者的连续有限时间一体化控制方法，同时实现多领航者姿态编队控制和多跟随者姿态合围控制；最后，搭建多航天器姿态模拟仿真平台并验证以上理论成果。项目研究成果有望为多航天器编队飞行的工程实践奠定坚实的理论和技术基础。

本项目针对航天器编队飞行任务中存在的实际工程问题及需求，深入开展了多航天器姿态协同控制方法的研究，致力于满足控制系统快速响应、精确收敛等性能需求。首先，针对具有转动惯量不确定性、外界扰动的多航天器系统，分别研究了静态和动态领航者下的有限时间姿态协同跟踪控制问题，提出了基于输出反馈的连续姿态协同控制方法；其次，在通信受限情况下，建立了系统实现跟踪控制与通信率、网络拓扑结构以及系统参数之间的确定关系，进一步，研究了状态受限下有限时间协同跟踪控制问题，设计了基于障碍李雅普诺夫函数的光滑控制方法；然后，针对姿态系统存在执行机构故障等因素，分别提出了有限时间和固定时间姿态容错控制方法；最后，研究了具有多领航者和多跟随者的姿态协同编队和合围一体化有限时间控制方法。以上结果共发表国际SCI期刊论文7篇和EI国际会议论文1篇，其中均发表于控制领域顶级SCI期刊，包括IEEE Transactions on Industrial Informatics、IEEE Transactions on Neural Networks and Learning Systems、IEEE Transactions on Cybernetics、International Journal of Robust and Nonlinear Control等。基于以上理论研究成果形成了一套较为完整的多航天器姿态协同控制方法，并在所搭建的姿态协同仿真平台上得到了验证，为大规模航天器编队飞行的工程实践提供了坚实理论和技术基础。