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。     

Intelligent modeling and identification of aircraft nonlinear flight

In this paper, a new approach has been proposed to identify and model the dynamics of a highly maneuverable fighter aircraft through artificial neural networks（ANNs）. In general, aircraft flight dynamics is considered as a nonlinear and coupled system whose modeling through ANNs, unlike classical approaches, does not require any aerodynamic or propulsion information and a few flight test data seem sufficient. In this study, for identification and modeling of the aircraft dynamics, two known structures of internal and external recurrent neural networks（RNNs） and a proposed structure called hybrid combined recurrent neural network have been used and compared.In order to improve the training process, an appropriate evolutionary method has been applied to simultaneously train and optimize the parameters of ANNs. In this research, it has been shown that six ANNs each with three inputs and one output, trained by flight test data, can model the dynamic behavior of the highly maneuverable aircraft with acceptable accuracy and without any priori knowledge about the system.     

A fault-tolerant neural aided controller for aircraft auto-landing

This paper presents a neural-aided controller that enhances the fault tolerant capabilities of a high performance fighter aircraft during the landing phase when subjected to severe winds and failures such as stuck control surfaces. The controller architecture uses a neural controller aiding an existing conventional controller. The neural controller uses a feedback error learning mechanism and employs a dynamic Radial Basis Function neural network called Extended Minimal Resource Allocating Network (EMRAN), which uses only on-line learning and does not need a priori training. The conventional controller is designed using a classical design approach to achieve the desired autonomous landing profile with tight touchdown dispersions called herein as the pillbox. This design is carried out for no failure conditions but with the aircraft being subjected to winds. The failure scenarios considered in this study are: (i) Single faults of either aileron or elevator stuck at certain deflections, and (ii) double fault cases where both the aileron and elevator are stuck at different deflections. Simulation studies indicate that the designed conventional controller has only a limited failure handling ability. However, neural controller augmentation considerably improves the ability to handle large faults and meet the strict touchdown dispersion requirements, thus enlarging the fault-tolerance envelope.     

战斗机非线性飞行控制技术的研究与发展

战斗机非线性飞行控制技术研究综述。首先介绍战斗机技术发展走向,然后对当前主要战斗机飞行控制设计方案进行分析,并重点介绍神经网络在飞行控制中的应用研究。最后指出,基于神经网络的智能控制方案作为飞行控制的重要研究方向,将为未来先进战斗机飞行控制系统设计提供重要的解决方案。     

基于神经网络的超机动飞机自适应重构控制

 讨论了一种基于神经网络的超机动飞机直接自适应重构控制方法。飞机的基本控制律采用非线性动态逆方法设计,对于模型不准确和舵面故障等因素导致的逆误差采用神经网络进行在线补偿。通过仿真表明,在飞机发生舵面故障时,神经网络通过自适应地补偿逆误差,可以快速在线重构控制律,保持飞机稳定和一定的操纵品质。     

三轴式无人旋翼飞行器及自适应飞行控制系统设计

设计了一种操控简便的三轴式无人旋翼飞行器,由三组共轴双旋翼组成,各旋翼由直流电机直接驱动,只需调节各电机转速就能控制旋翼飞行器运动姿态和轨迹。为使三轴式无人旋翼飞行器飞行控制系统设计得到有效验证,研究了旋翼飞行器的飞行动力学非线性建模,运用叶素动量理论建立了共轴双旋翼变转速旋翼载荷计算方法,分析了旋翼入流分布对共轴双旋翼气动载荷模型的影响,通过试验验证了共轴双旋翼气动载荷计算模型的正确性。由于旋翼飞行器飞行动力学模型的非线性及未建模动力学的影响,难于建立非常精确的数学模型,给飞行控制系统设计带来了挑战。本文根据旋翼飞行器飞行动力学非线性模型推导出了旋转动力学模型逆和平移动力学模型逆控制器,利用神经网络在线自适应修正模型逆误差,采用线性PD或PI控制器调节指令跟踪误差,应用由向心回转和垂直上升组合的机动科目进行了仿真验证,给出了具有外界阵风干扰模拟的仿真结果,表明所设计的飞行控制系统具有自适应性和鲁棒性,能实现精确的轨迹跟踪控制。     

Application of neural networks for the design of flight control algorithms. II Adaptive tuning of neural network control law

In this paper, we consider different approaches for the neural network controller tuning in the flight control system. Two of the most common tuning approaches in the adaptive control theory are applied. The first one uses parameter identification technique and consists in solving a real-time regression problem for the control law. The second approach is based on the Lyapunov direct method, which utilizes a tracking error as an absolute measure of tuning performance. The neural network control law are designed for the three-axis flight control problem and tested on the full nonlinear model of a fighter aircraft. Closed loop simulation results are presented and two adaptation algorithms are compared in the case of abrupt change of aircraft dynamics.     

MIMO变结构飞行重构控制系统设计

基于滑模变结构控制,提出一种MIMO(multiple input multiple output)的飞行重构控制系统的频域设计方法.将滑模变结构控制与飞行重构控制相结合,解决了飞行重构控制技术中故障检测和系统参数辨识的问题.引入渐近观测器和hedge模型增加重构控制系统对衍生未建模动态的鲁棒性;引入作动器模型、输出饱和限制和驾驶员模型,使变结构重构控制系统设计方法变得更为有效和实用;以某型飞机的横航向飞行控制系统为例,进行设计模拟.结果表明:在飞机气动参数大幅突变和操纵面严重受损的情况下,飞机仍能保持良好的性能.      

待机状态下机翼结冰的快速计算方法

飞机在飞行过程中遭遇结冰气象,尤其是过冷大水滴结冰条件所导致的机翼结冰,将严重影响飞机的气动性能与操纵品质,从而导致飞行故障或飞行事故,是飞机飞行安全的重要影响因素,是飞机研制中必须重点解决的难题。虽然针对结冰已建立有多种计算仿真程序,但快速获得结冰模拟冰形,包括过冷大水滴条件下的结冰模型是工程师们一直追求的解决方案。利用前向多层神经网络,尝试建立了一种针对待机状态下的机翼结冰模型。此方法基于坐标转换原理,将获取的采用直角坐标表达的标准翼型的机翼结冰数据转换为对应的极坐标表达的数据,以飞行参数、气象参数和极坐标角度作为输入,极坐标模值作为输出训练构建出此种机翼结冰模型。使用所建立的模型,仿真预测结果表明该计算方法具备快速性和精确性,其计算精度可满足实际要求。     

飞机短周期运动参数的神经网络辨识

本文提出了一种用馈神经网络辨识飞机短周期运动参数的新方法。该方法的特点是不需要预先假定模型，只需要用获得的飞行数据训练一前馈馈神经网络，就可以用这个网络直接辨识参数的值。模拟飞行数据的仿真结果证明所提出的方法是有效的并且具有较好的鲁棒性。     

Design of safety monitor schemes for a fault tolerant flight control system

For a research aircraft, "conventional" control laws (CLs) are implemented on a "baseline" flight computer (FC) while research CLs are typically housed on a dedicated research computer. Therefore, for an experimental aircraft used to test specific fault tolerant flight control systems, a safety logic scheme is needed to ensure a safe transition from conventional to research CLs (while at nominal conditions) as well as from research CLs at nominal conditions to conditions with "simulated" failures on specific control surfaces. This paper describes the design of such a safety scheme for the NASA Intelligent Flight Control System (IFCS) F-15 Program. The goals of the IFCS F-15 program are to investigate the performance of a set of fault tolerant CLs based on the use of dynamic inversion with neural augmentation. The different transitions are monitored using information relative to flight conditions and controller-related performance criteria. The testing of the scheme is performed with a Simulink-based flight simulation code and interface developed at West Virginia University for the NASA IFCS F-15 aircraft.     

Closed-loop dynamic control allocation for aircraft with multiple actuators

A closed-loop control allocation method is proposed for a class of aircraft with multiple actuators. Nonlinear dynamic inversion is used to design the baseline attitude controller and derive the desired moment increment. And a feedback loop for the moment increment produced by the deflections of actuators is added to the angular rate loop, then the error between the desired and actual moment increment is the input of the dynamic control allocation. Subsequently, the stability of the closed-loop dynamic control allocation system is analyzed in detail. Especially, the closedloop system stability is also analyzed in the presence of two types of actuator failures: loss of effectiveness and lock-in-place actuator failures, where a fault detection subsystem to identify the actuator failures is absent. Finally, the proposed method is applied to a canard rotor/wing (CRW) aircraft model in fixed-wing mode, which has multiple actuators for flight control. The nonlinear simulation demonstrates that this method can guarantee the stability and tracking performance whether the actuators are healthy or fail.     

基于RBF神经网络的飞参数据预处理

由于存在干扰,飞参系统记录的发动机参数中,经常会有不少间断点和奇异值.为了利用数据对发动机性能趋势进行预测,必须对数据进行预处理.发动机作为一个系统,其各主要输入和输出参数之间必然存在着一定的函数关系.本文研究了利用RBF神经网络和参数之间的关系对数据进行预处理,得到了较为正常的数据,结果表明该方法是有效的.     

Application of neural networks for synthesizing flight control algorithms. I neural network inverse dynamics method for aircraft flight control

The application of artificial neural networks for aircraft motion control, in particular, for creation of nonlinear algorithms of the aircraft remote control system (RCS) is considered. Aircraft as a control object is represented as a multidimensional nonlinear dynamic system and nonlinear control methods are used to operate this system. The control loop is constructed using the method of inverse dynamics based on the feedback linearization principle. The nonlinear control law is represented as a neural network being learned (adjusted) by recorded or incoming measurements of motion parameters. Synthesis and testing of neural network control algorithms is performed with the fully nonlinear mathematical model of a maneuverable aircraft for three control channels. Simulation results of the closed system are presented.     

Reconfigurable Flight Control Design for Combat Flying Wing with Multiple Control Surfaces

With control using redundant multiple control surface arrangement and large-deflection drag rudders,a combat flying wing has a higher probability for control surface failures.Therefore,its flight control system must be able to reconfigure after such failures.Considering three types of typical control surface failures(lock-in-place(LIP),loss-of-effectiveness(LOE) and float),flight control reconfiguration characteristic and capability of such aircraft types are analyzed.Because of the control surface redundancy,the aircraft using the dynamic inversion flight control law already has a control allocation block.In this paper,its flight control configuration during the above failures is achieved by modifying this block.It is shown that such a reconfigurable flight control design is valid,through numerical simulations of flight attitude control task.Results indicate that,in the circumstances of control surface failures with limited degree and the degradation of the flying quality level,a combat flying wing adopting this flight control reconfiguration approach based on control allocation could guarantee its flight safety and perform some flight combat missions.     

QFT force loop design for the aerodynamic load simulator

A dynamic load simulator which can reproduce on-ground the aerodynamic hinge moment of a control surface is an essential rig for conducting performance and stability tests of aircraft actuation systems. The hinge moment varies widely over the flight envelope, depending on the specific flight condition and maneuvering status. To replicate the wide spectrum of this hinge moment variation within some accuracy hounds, a force controller is designed based on the Quantitative Feedback Theory (QFT). A dynamic model of the load actuation system Is derived, and compared with experimental results. Through this comparison, a nominal model of the load actuation system with some uncertainty bounds is developed. The efficacy of the QFT force controller is verified by a numerical simulation, in which combined aircraft dynamics, flight control law and hydraulic actuation system dynamics of the aircraft control surface are considered     

Design,modeling, and control of morphing aircraft:A review

A morphing aircraft can adapt its configuration to suit different types of tasks, which is also an important requirement of Unmanned Aerial Vehicles(UAV). The successful development of an unmanned morphing aircraft involves three steps that determine its ability and intelligent: configuration design, dynamic modeling and flight control. This study conducts a comprehensive survey of morphing aircraft. First, the methods to design the configuration of a morphing aircraft are presented and analyzed...     

Security, Internet connectivity and aircraft data networks

Internet connectivity which was in experimental stages only a few years ago is a reality today. Current implementations allow passengers to access Internet for pleasure and in some cases secure VPN access is provided to corporate networks. Several researchers are looking at the possibility of the existence of a total of three networks: passenger network (PN); crew network (CRN); and the control network (CON). Researchers envision an architecture where these three networks will co-exist in an airplane. The available Internet connectivity can be utilized for transporting flight critical information like cockpit flight data recorder (CFDR) data, digital flight data recorder (DFDR) data, cockpit voice recorder (CVR) data, and controller pilot data link communication. In addition, the Internet connectivity could also be used for other safety mechanisms like video surveillance and remote control of the flight. Security is one of the major concerns that affect the successful deployment of aircraft data networks (ADN) and other safety features. Several studies have been carried out to secure the network using firewalls and intrusion detection system but so far no study has focused on securing the communication channel (between the aircraft and the ground station) and its impact on the ADN. The scope of this research is to determine the viability and need of a security mechanism. The research also focuses on the performance of different security architectures and determine their usability in the framework of an ADN.     

一种动态面神经网络鲁棒飞行控制律设计

提出一种基于RBF(Radial Based Function)神经网络和动态面控制的飞行控制律设计方法。针对飞机气动参数变化引起的非线性和不确定性,通过RBF神经网络在线逼近。动态面飞行控制律消除了反推设计方法中由于对虚拟控制反复求导而导致的复杂性问题。同时,在控制律设计中引入一个鲁棒因子项来补偿外界干扰和神经网络的逼近误差,提高了系统的鲁棒性,使整个系统获得更好的跟踪控制性能。基于Lyapunov稳定性定理证明了闭环系统的所有信号半全局一致终结有界,并且通过适当选择设计参数,跟踪误差可收敛到原点的一个小邻域内。最后,通过飞机俯仰运动飞行的数值仿真验证了该方法的有效性。     

Data communications concept for a SATS scenario

The small aircraft transportation system (SATS) concept envisions doorstop to destination transportation in a safe and timely manner. Data communications are a key component in achieving the aviation-related operational performance improvements that are sought. However, data communication doesn't start when you get into the aircraft; it starts back at the location where the flight is planned. In fact, data communications support the pilot in all phases of the flight: flight planning, pre-flight, departure, en route, transition, approach, landing, and rollout as well as for a missed approach. The Internet is being used to perform flight planning activities, and the mobile communications available today support Internet access en route to the departure airfield. On-board the aircraft, data communications provide surveillance and air traffic control (ATC) support to the pilot. The location of other aircraft is available to the pilot and ATC system through automatic dependent surveillance-broadcast (ADS-B) and traffic information service-broadcast (TIS-B) applications that transmit the location of other aircraft in the vicinity. Other aircraft locations are used to forecast potential conflicts and, thus, enhance flight safety. As the aircraft nears a SATS-equipped airfield, the pilot uses data link messages to request a landing sequence. The airport management module (AMM) provides a landing sequence assignment to the aircraft. As the pilot maneuvers the aircraft for a landing, he/she is using data-linked surveillance data to determine the location of other aircraft and maintain a safe separation distance between aircraft even in a low visibility environment.