【作者】 方炜；

【导师】 姜长生；

【作者基本信息】 南京航空航天大学 ， 控制理论与控制工程， 2008， 博士

【摘要】 空天飞行器(ASV)是各国正在大力发展的一种新型航空航天飞行器,具有很高的军事和民用价值。在其高超声速再入飞行过程中,ASV的大攻角再入使其速度、高度和姿态变化剧烈,呈现出强烈的非线性及耦合性等动态特性,并且受到大量的外界干扰和内部不确定的影响,这使得ASV的控制系统设计成为一项极具挑战的研究课题。围绕这一基础科学问题,本文对ASV再入飞行的高精度、强稳定、快速自适应预测控制问题展开了较为深入的研究。首先,建立和完善超声速和高超声速再入飞行条件下的ASV的质心运动方程和绕质心运动的6自由度动力学和运动学方程,开环分析表明了该模型能够反映ASV再入飞行的复杂非线性、耦合性以及快速时变性等特点,具有一定的代表性,可以满足未来ASV先进制导和控制等问题的理论研究和仿真验证需要。其次,利用一种有限时间预测控制方法,考虑了高超声速飞行条件下的ASV再入制导轨迹控制问题。结合ASV再入飞行中的约束条件,建立了飞行器的再入走廊,选择适当的阻力加速度信号作为飞行器的参考轨迹,然后基于泰勒展开设计了ASV再入制导轨迹预测控制系统,仿真结果表明了ASV制导控制系统的有效性。随后,针对具有不确定和复杂非线性等特性的ASV六自由度动力学方程,本文提出了基于模糊自适应重构系统的多输入多输出不确定非线性系统的预测控制律设计,并应用于解决空天飞行器模型不完全已知时的自适应姿态控制系统设计。通过模糊权值的在线调整设计出自适应模糊系统逼近被控对象中的未知函数,从而建立基于模糊重构系统的非线性预测控制系统。为了降低设计方法的保守性,根据系统的跟踪误差调整鲁棒控制器中的自适应参数,所设计的控制器保证了闭环系统所有信号一致最终有界。最后在ASV高超声速飞行条件下进行仿真验证,以检验该控制系统的有效性和鲁棒性。接着,进一步考虑到ASV再入飞行过程中所受的内部不确定和外界干扰的影响,将ASV的气动参数摄动和外界未知扰动统一看成系统的复合干扰,在理论分析的指导下提出一种基于补偿原理的非线性模糊预测控制新策略,通过模糊系统对复合误差进行了在线逼近,并将其代入控制器设计,从而抵消了复合干扰对整个闭环系统的影响,保证了非线性预测控制律的控制性能。同时,在模糊系统中增加“可变论域”的思想,根据系统的跟踪误差自动调整模糊论域的大小,以生成任意多条模糊规则,使得模糊系统的响应品质大大提高。理论分析表明高精度地逼近系统中存在的不确定,可大大提高控制器的控制效果,为此本文在前面研究的基础上结合干扰观测器技术,设计了基于模糊干扰观测器的非线性鲁棒预测控制方法,为解决非线性不确定系统的控制问题提供了新的途径。该设计方法充分利用了被控系统的有用信息,并结合模糊系统实现了对ASV复合干扰更为有效地逼近和控制,从而获得了更好的控制性能。基于Lyapunov稳定原理,本文给出了模糊干扰观测器的自适应调整律,并严格证明了系统的跟踪误差和干扰观测误差一致最终有界。最后,本文结合了滑模控制方法和预测控制方法,将滑模面引入非线性预测控制的二次型代数指标中,并利用代数指标的最优化设计滑模控制律,从而提出了一种基于预测滑模控制的非线性鲁棒自适应控制,使得控制器既有滑模控制鲁棒性强的特点,也具有连续预测控制的优化特点。同时由于模糊自适应系统的引入,使得控制器无需知道系统不确定的上界,从而大大减弱了滑模控制的抖振现象。更多还原

【Abstract】 Aerospace vehicle (ASV),which is the next generation reusable flying vehicle, has very important martial value and civil value. During its hypersonic re-entry flying, big angle of attrack (AOA) bring the fast change of ASV’s speed, attitude and attitude, its equation subjects to serious nonlinear and coupling characteristic. Moreover, ASV subjects to the unkown disturbancea and uncertainties, so the design of fly control system of ASV is an attractive subject in the modem control area. In this dissertation, the adaptive predictive control methods of MIMO uncertain system are designed and used to solve the control problems involving in the fly control system of ASV.First of all, a simulation model of an ASV is presented, which includes the centroid motion of ASV and a six degree-of-freedom kinetic equations and motion equations. Its open-loop dynamics and stability characteristics demonstrate that the proposed model can behave the intricate nolinear, coupling and fast time-varying of ASV’s re-entry, so it can be used in the following theory research and evaluation of advanced guidance and control methods.Secondly, a kind of finite horizon predictive method is used to design the re-entry guidance control law of ASV under hypersonic fly condition. With the restrictions of ASV, the re-entry corridor is defined. By choosing an appropriate drag acclertaion as the reference trajectory, the ASV’s guidance predictive law is designed by talyor series expansion. Simulation results of the ASV guidance system indicate the acceptable performance of the designed law to track the reference signal.Thirdly, according to the complex nonlinear and uncertain ASV kinetics motion, a class of MIMO uncertain nonlinear predictive control shceme is investigated based on fuzzy reconfigurable systems, and its valid application on the robust adaptive flight control system of ASV with unkown model. By adjusting the fuzzy weight parameters on-line, the fuzzy system is used to approximate the unkown function and construct the nonlinear predictive control law. Furthermore, on the purpose of alleviating the conservatism of controller, an adaptive robust controller is constructed to compensate approximation error of fuzzy system. Using Lyapunov theory, it has been proven that the overall system were uniformly ultimately bounded. Finally, the flight control system of ASV under hypersonic condition is designed by the proposed method. Simulation results demonstrate the effectiveness of the method.In the following, the inner uncertainty and unkown disturbance of ASV are considered, which are regarded as composite disturbance. Under the guidance of theoretical analysis, a novel predictive control structure is developed by combining the current predictive method with the well-known compensation idea, the composite disturbance is compensated and elimeated by fuzzy systems, so the performance of the nonlinear predictive controller is guarraneted . Moreover, the concept of“variable universe”is introduced into fuzzy system, according to the tracking error, the contraction and expansion factor of variable universe fuzzy controller is adjusting, which lead to the improvement of response quality of fuzzy system.Theoretical analysis illustrates that the accurate approximation to the uncertainty can enchance the performance of the controller greatly. Combining with the previous research, this thesis investigates a class of nonlinear robust predictive control scheme based on fuzzy disturbance observer, the scheme can be provided as an effective method to a kind of nonlinear uncertain systems. Using the useful information of the controlled system, the method realized the accurate approximation and control, so the performace of controller is enhanced. By Lyapunov stability theory, the adjusting law of fuzzy disturbance observer is given, moreover, the ultimate boundness of the tracking error and oberserve error has been proven.Finally, considering the slide mode control and predictive control, this article puts the sliding surface into the quadratic index, and gets the sliding mode control law through the minimum of the index, so the nonlinear robust adaptive controller is constructed which is based on predictive sliding mode. The proposed scheme is highly robust as slide mode control, and has the virture of optimization as contious predictive control. Applying the fuzzy adaptive system to the controller, the prior knowledge about the upper bound of uncertainty can be removed, so the chattering of sliding mode is weakened greatly.更多还原