基于神经网络的无人直升机姿态控制系统设计

首先根据模型参考自适应控制理论,将模型逆与在线神经网络结合,设计了神经网络自适应姿态控制系统。接着叙述反馈线性化及模型逆理论,分析系统的模型跟踪误差动力特性,设计神经网络控制器及在线算法。然后以某无人直升机俯仰通道为例,对神经网络姿态控制系统进行仿真。结果表明该系统能够对未建模特性、参数不确定性等引起的模型逆误差进行自适应,而且在传感器输出中具有白噪声时仍然能够获得较好的响应特性。     

基于滑模神经网络的自主飞艇姿态控制

针对自主飞艇飞行环境的不确定性,提出了一种基于自适应滑模神经网络的姿态控制系统.平流层高空飞行环境对飞艇控制产生了许多不确定性因素,利用自适应变结构控制和神经网络方法设计了飞艇的俯仰通道控制器.非线性仿真结果表明:控制器能够适应对象结构参数及外部扰动的大范围变化,满足姿态控制稳定性要求,同时也消除了变结构控制系统的抖振,具有良好的鲁棒性和动态性能.     

一类非仿射输入和不确定性系统的输出跟踪控制

针对一类具有非仿射输入和不确定性的混沌系统,提出了鲁棒自适应RBFNN反演控制。RBF神经网络用来逼近设计过程中所有的未知非线性函数,并且对逼近的误差设计了鲁棒项,设计的自适应律是针对RBF所有权值的上界而不是权值本身,因而在线自适应参数数量减少了,节约了计算时间。该方法保证了混沌系统的输出能跟踪任意参考信号,并且跟踪误差是一致渐进收敛的。仿真结果表明该方法的可行性和有效性。     

超低空空投航迹倾角自适应跟踪控制

针对超低空空投下滑阶段考虑执行器输入死区、不确定性大气扰动以及模型存在未知非线性等因素干扰轨迹精确跟踪等问题,提出了一种自适应神经网络动态面跟踪控制方法.建立了含执行器输入死区的超低空空投载机纵向非线性模型,采用神经网络逼近模型中未知非线性函数,引入非线性鲁棒补偿项消除了执行器死区建模误差和外界扰动.应用Lyapunov稳定性理论证明了闭环系统所有信号均是有界收敛的.仿真验证表明,所提方法既保证了轨迹跟踪的精确性,又具有强鲁棒性.     

基于神经网络的MIMO非线性最小相位系统鲁棒自适应控制

 针对一类模型未知的具有不确定性和外部干扰的多输入多输出（MIMO）非线性最小相位系统提出了鲁棒自适应输出反馈跟踪控制方案。用高斯径向基函数（RBF）神经网络逼近对象未知非线性,用高增益观测器估计系统不可测量状态。所设计的鲁棒自适应控制器不仅能使闭环系统稳定,所有状态有界,而且跟踪误差一致最终有界,并保证最终边界足够小。仿真结果表明了所提出方法的有效性。     

基于模糊模型的鲁棒自适应重构飞行控制

提出了一种基于模糊模型的歼击机鲁棒自适应重构控制方案。整个控制方案基于T S模糊模型,将歼击机各飞行状态的局部线性调节器与鲁棒自适应神经网络重构控制器相结合,避免了传统的增益预置方法中控制律在不同工作点之间切换造成的参数突变对系统性能的影响,可以保证系统在全局上拥有局部工作点具有的期望性能,证明了重构系统的全局闭环渐近稳定性。所提出的带有补偿项的完全自适应RBF神经网络,通过在线自适应调整RBF神经网络的权重、函数中心和宽度,提高了神经网络的学习能力,同时通过自适应补偿项来在线估计神经网络的近似误差边界,可以有效地在线修正建模误差、外扰及操纵面故障等因素的影响,保证系统的操纵品质。仿真结果表明了所提出方法的有效性。     

基于RBF神经网络的机器人鲁棒自适应控制器设计

在考虑机器人系统中存在的模型不确定性的情况下,提出了一种基于RBF神经网络和反演技术的鲁棒自适应控制器的设计方法。首先,通过状态变换将机器人的模型转换为适用于反演技术的形式;然后,利用反演技术设计了鲁棒自适应控制器,用两个RBF神经网络分别对模型的不确定性进行了处理,并用Lyapunov稳定性理论推导出RBF神经网络的权重矩阵调节律以及相关的鲁棒项,证明了系统的全局稳定性;最后,进行了相应的仿真研究,验证了设计的正确性和有效性。     

基于神经网络自适应动态逆的结冰飞机飞行安全边界保护方法

考虑到飞机带冰飞行的安全问题，对结冰飞机进行安全边界保护成为一种有效的解决手段。基于神经网络自适应动态逆跟踪性能好、鲁棒性强的优点，提出了以安全关键飞行参数限制值作为神经网络自适应动态逆的输入，获取可用舵面偏转角的边界保护方法。建立了飞机本体动力学模型，采用高精度的数值模拟方法获得结冰数据库。设计了神经网络自适应动态逆控制律，通过在动态逆环节引入单隐层神经网络，对不确定性逆误差进行自适应补偿，增强了控制系统的鲁棒性。以俯仰姿态保持模式为例设计了结冰飞行闭环安全边界保护系统。以结冰飞机最小平飞速度的估算值作为飞机最低飞行速度，设计自动油门控制系统，实现对飞行速度的保护。通过仿真验证了设计的控制律具有较强的鲁棒性。对结冰严重程度线性增加情形下飞机状态参数的动态响应进行了分析。仿真结果表明，所设计的结冰边界保护系统，能够实现飞机在容冰飞行过程中对安全关键参数如迎角、飞行速度的实时保护。     

航空发动机神经网络自适应控制研究

本文研究神经网络自适应控制方法及其在航空发动机控制中应用。结合某型航空涡喷发动机,首先研究采用神经网络进行非线性动态系统辨识,包括神经网络模型辨识的格式、输入信号形式等问题。然后,提出了一种神经网络自适应控制方法,阐明了该方法基本结构、原理。最后,在选定的设计点处进行发动机控制系统设计,当偏离设计点时,利用神经网络很强的学习、适应能力,通过在线修正神经网络参数,使控制系统仍保持良好性能。      

RFNN control for PMLSM drive via backstepping technique

A robust fuzzy neural network (RFNN) control system is proposed in this study to control the position of the mover of a permanent magnet linear synchronous motor (PMLSM) drive system to track periodic reference trajectories. First, an ideal feedback linearization control law is designed based on the backstepping technique. Then, a fuzzy neural network (FNN) controller is designed to be the main tracking controller of the proposed RFNN control system to mimic an ideal feedback linearization control law, and a robust controller is proposed to confront the shortcoming of the FNN controller. Moreover, to relax the requirement for the bound of uncertainty term, which comprises a minimum approximation error, optimal parameter vectors and higher order terms in Taylor series, an adaptive bound estimation is investigated where a simple adaptive algorithm is utilized to estimate the bound of uncertainty. Furthermore, the simulated and experimental results due to periodic reference trajectories demonstrate that the dynamic behaviors of the proposed control systems are robust with regard to uncertainties.     

输入受限的吸气式高超声速飞行器自适应Terminal滑模控制

 针对吸气式高超声速飞行器的纵向运动模型,研究了输入受限时控制系统的设计问题,提出了一种内外环相结合的自适应Terminal滑模控制方法.以迎角和俯仰角速度作为内环控制对象,考虑到气动弹性模态和外部干扰,采用反步设计方法,设计出升降舵的自适应Terminal滑模控制律;以飞行速度作为外环控制对象,采用动态逆设计方法,设计出输入燃料当量比的动态逆自适应控制律,同时利用多层神经网络逼近控制律的饱和特性.基于Lyapunov稳定性理论,证明了采用此控制策略可以保证闭环控制系统的所有信号都是指数收敛至系统原点的一个有界邻域内,同时飞行器的气动弹性模态都是渐近稳定的.仿真结果表明,此控制策略能有效地处理吸气式高超声速飞行器在纵向运动过程中控制输入饱和受限的约束,在完成控制目标的同时,具有良好的过渡过程品质.     

基于径向基神经网络干扰观测器的空天飞行器自适应轨迹线性化控制

 研究了一种自适应轨迹线性化控制策略并应用于空天飞行器(ASV)飞行控制系统设计。通过理论分析指明当前轨迹线性化控制方法(TLC)对系统中的不确定存在鲁棒性不足的问题。为了解决这一问题，首先研究了一种径向基神经网络干扰观测器(RDO)技术，严格证明了RDO对于系统中不确定因素具有良好的逼近能力。然后利用RDO输出得到一种新的基于RDO的自适应TLC控制策略。神经网络自适应律采用Lyapunov方法设计，保证了闭环系统所有信号有界。最后采用新方案实现了ASV飞控系统，仿真结果表明整个闭环系统在鲁棒性能方面得到很大提高。     

基于神经网络的自动着陆飞行鲁棒自适应逆控制

针对飞机自动着陆飞行提出了基于神经网络的鲁棒自适应非线性动态逆控制器设计方案。首先采用非线性动态逆方法设计着陆飞行的基本控制律,再利用多层感知器神经网络设计适当的权值调整规则使其能够自适应地逼近和补偿逆误差。仿真结果表明,所设计的飞行控制系统是有效的,系统能够克服动态逆误差对着陆飞行控制带来的不利影响。     

Nonlinear Adaptive Robust Force Control of Hydraulic Load Simulator

This paper deals with the high performance force control of hydraulic load simulator. Many previous works for hydraulic force control are based on their linearization equations, but hydraulic inherent nonlinear properties and uncertainties make the conventional feedback proportional-integral-derivative control not yield to high-performance requirements. In this paper, a nonlinear system model is derived and linear parameterization is made for adaptive control. Then a discontinuous projection-based nonlinear adaptive robust force controller is developed for hydraulic load simulator. The proposed controller constructs an asymptotically stable adaptive controller and adaptation laws, which can compensate for the system nonlinearities and uncertain parameters. Meanwhile a well-designed robust controller is also developed to cope with the hydraulic system uncertain nonlinearities. The controller achieves a guaranteed transient performance and final tracking accuracy in the presence of both parametric uncertainties and uncertain nonlinearities; in the absence of uncertain nonlinearities, the scheme also achieves asymptotic tracking performance. Simulation and experiment comparative results are obtained to verify the high-performance nature of the proposed control strategy and the tracking accuracy is greatly improved.     

Adaptive formation control of quadrotor unmanned aerial vehicles with bounded control thrust

In this paper, the flight formation control problem of a group of quadrotor unmanned aerial vehicles (UAVs) with parametric uncertainties and external disturbances is studied. Unit-quaternions are used to represent the attitudes of the quadrotor UAVs. Separating the model into a translational subsystem and a rotational subsystem, an intermediary control input is introduced to track a desired velocity and extract desired orientations. Then considering the internal parametric uncertainties and external disturbances of the quadrotor UAVs, the priori-bounded intermediary adaptive control input is designed for velocity tracking and formation keeping, by which the bounded control thrust and the desired orientation can be extracted. Thereafter, an adaptive con-trol torque input is designed for the rotational subsystem to track the desired orientation. With the proposed control scheme, the desired velocity is tracked and a desired formation shape is built up. Global stability of the closed-loop system is proven via Lyapunov-based stability analysis. Numer-ical simulation results are presented to illustrate the effectiveness of the proposed control scheme.     

高速开关阀控电液位置伺服系统自适应鲁棒控制

为研究高精度的液压缸位置跟踪控制问题，设计了高速开关阀和换向阀组合控制液压缸的结构方案，并通过实验分析了高速开关阀的静态流量特性。建立了液压缸的连续可微摩擦模型，利用粒子群优化算法对其参数进行辨识。建立了系统的非线性数学模型，基于非连续参数映射和反步法设计了直接自适应鲁棒控制器，通过参数在线自适应调节来更新估计值和鲁棒反馈项支配参数不确定性，实验结果表明:在跟踪幅值为5mm，频率为0.4Hz的正弦信号时，最后一个周期的最大跟踪误差、平均跟踪误差及其标准差分别为0.638、0.25mm和0.405mm，与传统PID控制器相比，控制精度显著提升，旨在为实现高精度的数字阀控位置伺服技术提供有价值的参考。      

Decentralized adaptive sliding mode control of a space robot actuated by control moment gyroscopes

An adaptive sliding mode control(ASMC) law is proposed in decentralized scheme for trajectory tracking control of a new concept space robot.Each joint of the system is a free ball joint capable of rotating with three degrees of freedom(DOF).A cluster of control moment gyroscopes(CMGs) is mounted on each link and the base to actuate the system.The modified Rodrigues parameters(MRPs) are employed to describe the angular displacements,and the equations of motion are derived using Kane's equations.The controller for each link or the base is designed separately in decentralized scheme.The unknown disturbances,inertia parameter uncertainties and nonlinear uncertainties are classified as a ‘‘lumped" matched uncertainty with unknown upper bound,and a continuous sliding mode control(SMC) law is proposed,in which the control gain is tuned by the improved adaptation laws for the upper bound on norm of the uncertainty.A general amplification function is designed and incorporated in the adaptation laws to reduce the control error without conspicuously increasing the magnitude of the control input.Uniformly ultimate boundedness of the closed loop system is proved by Lyapunov's method.Simulation results based on a three-link system verify the effectiveness of the proposed controller.     

空天飞行器的鲁棒自适应模糊轨迹线性化控制

提出了鲁棒自适应模糊轨迹线性化控制(RAFTLC)方法并应用于空天飞行器(ASV)飞行控制系统设计.根据系统的先验知识, 设计出标称模糊系统对系统的未知干扰和不确定进行估计, 并通过鲁棒自适应控制项来克服标称模糊系统逼近误差和权值误差的影响.标称模糊系统逼近误差和权值误差的界在线调整.采用Lyapunov方法证明了闭环系统的所有信号一致最终有界.最后利用提出的控制方案设计了ASV飞行控制系统, 并在高超声速条件下进行了仿真验证.仿真结果表明了控制方案的有效性和鲁棒性.      

飞行/推进系统自适应神经网络综合控制仿真研究

 提出一种基于发动机喘振裕度自适应的飞行 /推进系统综合控制。在发动机喘振裕度较大的某些飞行条件或飞行包线内,通过调整喷口面积,使发动机喘振裕度保持在一个较小值,既保证发动机稳定工作,又增加发动机推力,从而改善飞机的性能。采用分散控制方案,综合控制系统由 5个控制子系统组成。各控制子系统的设计采用自适应控制和神经网络相结合的方法,所提出的参数和权重的自适应调整律保证系统的稳定性。全包线范围内飞机平飞加速和爬升数字仿真结果表明,该综合控制方法可缩短飞机的平飞加速时间和爬升时间。     

Polynomial networks based adaptive attitude tracking control for NSVs with input constraints and stochastic noises

This paper proposes a backstepping technique and Multi-dimensional Taylor Polynomial Networks (MTPN) based adaptive attitude tracking control strategy for Near Space Vehicles (NSVs) subjected to input constraints and stochastic input noises. Firstly, considering the control input has stochastic noises, and the attitude motion dynamical model of the NSVs is actually modeled as the Multi-Input Multi-Output (MIMO) stochastic nonlinear system form. Furthermore, the MTPN is used to estimate the unknown system uncertainties, and an auxiliary system is designed to compensate the influence of the saturation control input. Then, by using backstepping method and the output of the auxiliary system, a MTPN-based robust adaptive attitude control approach is proposed for the NSVs with saturation input nonlinearity, stochastic input noises, and system uncertainties. Stochastic Lyapunov stability theory is utilized to analysis the stability in the sense of probability of the entire closed-loop system. Additionally, by selecting appropriate parameters, the tracking errors will converge to a small neighborhood with a tunable radius. Finally, the numerical simulation results of the NSVs attitude motion show the satisfactory flight control performance under the proposed tracking control strategy.