旋翼式无人机编队大机动抗风扰鲁棒控制

Flight formations of multiple unmanned rotorcrafts have wide potential applications in cooperative search and combat. The dynamics of this class of unmanned aerial vehicles (UAVs) involves high nonlinearity and large uncertainties, especially in aggressive maneuvers. This project investigates the robust formation control problems for unmanned quadrotors, tail-sitters, and regular unmanned helicopters under aggressive maneuvers. .In this project, a robust control method will be firstly proposed for this kind of UAVs to track aggressive trajectory references and restrain the effects of multiple uncertainties including the wind gusts on the closed-loop control system. Moreover, the feedback linearization technique will be introduced and a robust nonlinear formation control method will be developed for a group of homogeneous unmanned rotorcrafts under wind gusts. Furthermore, the communication delays are considered and a formation protocol will be developed for a team of homogeneous unmanned rotorcrafts subject to wind gusts. Last, a robust formation control law based on the reinforcement learning theory will be proposed for a set of heterogeneous unmanned rotorcrafts with partially or completely unknown dynamics. This project includes dynamical system analysis, flight control law design, and experimental verification. The project provides the theoretical basis for the formation control of multiple unmanned rotorcrafts under aggressive maneuvers and wind gusts in practical applications.

旋翼式无人机编队在协同搜索与作战领域具有广泛的应用前景。该类无人机具有非线性程度高、不确定性大等特点，尤其在做大机动飞行时。本项目拟以四旋翼式、尾座式无人机和常规布局无人直升机为研究对象，研究其大机动编队飞行时鲁棒抗风扰控制问题。. 针对该类无人机，本项目拟首先提出一种单机的鲁棒轨迹跟踪控制方法，使其能实现对大机动轨迹参考信号的跟踪，并抑制外界风扰等不确定性的影响。其次，拟引入反馈线性化技术并提出一种多架同构无人机鲁棒非线性抗风扰编队控制方法。再次，针对无人机之间通信存在延时的情况，拟提出一种多架同构无人机鲁棒编队控制协议。最后，考虑这类无人机部分动态或全部动态未知时，拟引入强化学习理论设计多架异构无人机鲁棒编队控制律。研究工作将围绕动态系统分析、飞行控制律设计、实验验证等环节展开。研究成果将为旋翼式无人机大机动抗风扰编队飞行的工程实践提供理论基础。

旋翼式无人机编队大机动抗风扰鲁棒控制是无人机自主控制研究中备受关注的科学和工程问题，针对复杂多输入多输出欠驱动旋翼无人机系统的非线性和耦合程度高、参数不确定性大、易受外界风扰、未知动态以及编队系统通信延时等问题，研究多旋翼式无人机鲁棒编队控制原理和方法具有重要意义，在提高多无人机系统对于外部风扰鲁棒性的同时，使之具有快速突防等能力，实现对大机动移动目标的协同攻击。本项目以多种旋翼式无人机为研究对象，针对这类问题进行了深入的研究：首先提出了一种单机旋翼式无人机的鲁棒轨迹跟踪控制方法，实现了对大机动轨迹参考信号的跟踪，并抑制外界风扰等不确定性的影响；其次，基于反馈线性化技术，提出了一种多架同构无人机鲁棒非线性抗风扰编队控制方法，抑制了非线性动态、参数不确定性和外界风扰的影响，同时降低了鲁棒控制器设计的保守性；再次，提出了一种无人机间通信延时情况下多架同构无人机鲁棒编队控制协议，抑制了通信延时对于闭环控制系统的影响；最后，基于强化学习理论，提出了一种数据驱动的多架异构无人机鲁棒编队控制律，解决了在旋翼式无人机部分动态或全部动态未知情况下的鲁棒编队控制问题。通过本项目的研究，为该旋翼式无人机在不同飞行条件和任务下，提供了有效的控制系统设计和研究的新方法，为该类无人机大机动抗风扰编队飞行的工程实践提供了理论基础。