基于自适应驾驶员最优控制模型的飞行品质预测

传统飞机驾驶员最优控制模型采用卡尔曼滤波,无法反映飞行员对未知环境的适应能力,在飞行试验中有时存在与飞行员实际评分不一致的现象.为此,采用自适应状态估计理论对传统驾驶员最优控制模型进行了修正,提出了基于自适应飞机驾驶员最优控制模型的飞行品质评估方法.通过对比飞行试验和模型仿真结果表明了这一评估方法的可行性,所采用的修正加权系数得到的评分结果精度更高.研究结果表明,飞行员评分与指标函数加权系数比值相关,采用变化加权系数比值得到的评估结果与飞行员实际评分更为吻合.随着飞机动态特性的变差,这一比值将不断增大,飞行员将投入更多精力进行飞行状态监测,进而导致飞行员降低对飞行品质的主观评价.     

舰载机飞行员驾驶技术与飞行安全研究

针对无法将驾驶技术和飞行安全建立有效定量关联的现状,基于自适应飞机驾驶员最优控制模型,应用飞行员着舰多层次模糊综合评价方法,通过设定不同的飞行员模型参数,参照着舰技能评价体系,建立了飞行员着舰技能样本库.通过多元非线性回归方法,建立了飞行员技能的多元非线性回归模型,并通过剩余标准差和样本比对法证明了这一模型的准确性.可以通过测定上述飞行员参数,运用该飞行技能回归模型,为飞行员选拔提供参考.     

基于相似构型决策的舰载机驾驶员建模与评估

舰船运动和舰尾流等复杂海况是影响舰载机着舰难易度和精确度的重要因素，建立驾驶员着舰控制行为模型并构建人机闭环系统是评估复杂海况对驾驶员操纵负荷影响的有效手段。此外，新一代舰载机存在多种构型配置，并且即便是同一飞机其每次开始着舰时也可能处于不同的质量和气动构型下，因此在驾驶员建模中有必要考虑飞机构型变化。针对舰载机着舰操纵负荷评估问题，基于飞行试验手段和频域计算方法，建立了纵向俯仰跟踪的Hess结构驾驶员模型。考虑飞机着舰构型的变化，基于CAP准则和控制扩展方法，构建了舰载机典型构型库，并应用驾驶员在环仿真试验建立了与各飞机构型匹配的驾驶员模型，形成驾驶员模型库。在此基础上，给出了相似构型决策准则和方法，提出了针对指定飞机构型的驾驶员行为模型预测方法。针对舰载机在复杂海况下的着舰任务开展仿真分析，结果表明，本文方法能够支撑研究不同舰载机构型的操纵负荷的影响。     

基于光学引导下飞行员操纵着舰建模

针对驾驶员操纵舰载机着舰的任务,建立一个在菲涅尔透镜光学助降系统(FLOLS)下的改进的驾驶员模型。在人工着舰时,飞行员驾驶飞机对下滑轨迹跟踪为基础,对驾驶员控制飞机的增益进行自适应控制的改进,对驾驶员在FLOLS光学引导下的视觉感受模型进行模糊控制的改进。结合舰载机动力学特性对舰载机着舰任务进行仿真。仿真结果表明了模型的合理性,可为飞行员操纵着舰提供参考。     

一种飞行员最优控制模型加权系数计算方法

目前飞行员最优控制模型(OCM)指标函数加权系数没有统一的选取方法,在计算中大多由试验数据拟合或试凑决定,由此带来了方法使用的随意性和飞行品质预测结果的不一致性。针对这一问题,提出了一种以飞行员最优注意力分配假设为基础的OCM加权系数计算方法。通过数学仿真对注意力最优分配假设和加权系数计算方法进行了验证,仿真结果证明了假设的合理性和加权系数计算方法的可行性。     

基于有向图的航天器编队飞行自适应姿态协同控制

在有向通信拓扑下研究了编队航天器自适应姿态协同控制问题。针对航天器编队飞行系统中存在外部扰动和模型不确定性的情况,通过选取包含相对姿态误差和绝对姿态误差的辅助变量,提出了一种鲁棒自适应控制策略。提出了自适应律估计转动惯量矩阵和扰动上界等未知参数,并且利用Lyapunov稳定性理论分析了闭环系统的渐近稳定性。与滑模控制等传统鲁棒控制不同,所设计的鲁棒自适应控制器是连续的,更便于航天器编队飞行系统的实现。最后通过仿真验证了该控制策略能够实现高精度的编队飞行跟踪控制。     

舰载机着舰任务下的飞行员建模研究

为研究舰载机在下滑着舰时飞行员的操纵策略,建立了着舰任务下的双通道飞行员模型.以人工着舰时飞行员控制飞机进行轨迹跟踪为背景,以着舰时惯用的“反区”操纵技术为依据,确立了舰载机飞行员控制律,并对控制增益参数进行了自适应设计.最后结合F-18A舰载机,运用SIMILINK软件进行了仿真.结果表明,该模型能很好地反映着舰任务下的飞行员操纵特性,可为飞行员着舰操纵提供参考.     

一种飞行员仿真系统的设计

分析了飞行仿真领域中对飞行员仿真的需求,提出了"全飞行周期飞行员仿真系统"的设想。描述了该系统的功能、接口和配置。针对飞行员行为建模问题提出了一种具有四个层次的模型框架用来组织飞行员在驾驶飞机时的各种简单或复杂的行为模型,此框架可作为飞行员仿真系统的软件框架。给出了人-机闭环仿真系统的着舰过程计算结果。     

舰载飞机着舰下滑跃升操纵策略仿真与分析

为了对舰载飞机着舰下滑时的跃升操纵策略进行研究,建立了飞机飞行动力学模型和驾驶员模型.在此基础上,根据“50 ft跃升规范”要求,以F/A-18舰载飞机为例,运用MATLAB/SIMULINK软件,对采取不同操纵策略下的飞机跃升飞行历程进行了仿真分析.结果表明,F/A-18舰载飞机在着舰下滑状态下无法达到“50 ft跃升规范”.此外,飞机在不开启动力补偿的情况下,飞行员需同时操纵油门杆和驾驶杆来实现轨迹跃升.     

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.      

Augmented flight dynamics model for pilot workload evaluation in tilt-rotor aircraft optimal landing procedure after one engine failure

An augmented flight dynamics model is developed to extend the existing flight dynamics model of tilt-rotor aircraft for optimal landing procedure analysis in the event of one engine failure.Compared with the existing flight dynamics model, the augmented model involves with more pilot control information in cockpit and is validated against the flight test data. Based on the augmented flight dynamics model, the optimal landing procedure of XV-15 tilt-rotor aircraft after one engine failure is formulated into a Nonlinear Optimal Control Problem(NOCP), solved by collocation and numerical optimization method. The time histories of pilot controls in cockpit during the optimal landing procedure are obtained for the evaluation of pilot workload. An evaluation method which can synthetically quantify the pilot workload in time and frequency domains is proposed with metrics of aggressiveness and cutoff frequencies of pilot controls. The scale of the pilot workload is compared with those of the shipboard landing procedures, bob-up/bob-down and dash/quickstop maneuvers of UH-60 helicopter. The results show that the aggressiveness of pilot collective and longitudinal controls for the tilt-rotor aircraft optimal landing procedure after one engine failure are higher than those for UH-60 helicopter shipboard landing procedures up to the condition of sea state 4, while the pilot cutoff frequency of collective control is lower than that of the bob-up/bob-down maneuver but the pilot cutoff frequency of longitudinal control is higher than that of the dash/quick-stop maneuver. The evaluated pilot workload level is between Cooper–Harper HQR Level 2 and Level 3.     

Prediction of pilot workload in helicopter landing after one engine failure

This paper presents a method to predict the pilot workload in helicopter landing after one engine failure. The landing procedure is simulated numerically via applying nonlinear optimal control method in the form of performance index, path constraints and boundary conditions based on an augmented six-degree-of-freedom rigid-body flight dynamics model, solved by collocation and numerical optimization method. UH-60A helicopter is taken as the sample for the demonstration of landing after one engine failure. The numerical simulation was conducted to find the trajectory of helicopter and the controls from pilot for landing after one engine failure with different performance index considering the factor of pilot workload. The reasonable performance index and corresponding landing trajectory and controls are obtained by making a comparison with those from the flight test data. Furthermore, the pilot workload is evaluated based on wavelet transform analysis of the pilot control activities. The workloads of pilot control activities for collective control, longitudinal and lateral cyclic controls and pedal control during the helicopter landing after one engine failure are examined and compared with those of flight test. The results show that when the performance index considers the factor of pilot workload properly, the characteristics of amplitudes and constituent frequencies of pilot control inputs in the optimal solution are consistent with those of the pilot control inputs in the flight test. Therefore, the proposed method provides a tool of predicting the pilot workload in helicopter landing after one engine failure.     

基于模糊推理的舰载机进舰过程安全性仿真分析

 舰载机进舰过程复杂,驾驶员对舰载机的控制是影响进舰安全的重要因素,驾驶员操作的偏差受到人员自身状态、环境状态以及设备(舰载机)状态的综合影响。以国外某型舰载机为例,提出了基于模糊推理的舰载机进舰过程安全性建模与仿真方法。考虑舰载机进舰过程中驾驶员控制的不确定性,分析了引起驾驶员不确定性的人-机-环因素,基于模糊推理建立不确定因素引起的驾驶员控制偏差模型,并在此基础上建立了舰载机进舰过程安全性仿真模型。通过仿真分析明确危险模式、危险程度以及危险发生的时刻点等关键因素的影响程度,为进一步保证舰载机进舰安全提供基础。     

推力矢量型V/STOL飞行器动态过渡过程的操纵策略优化

针对推力矢量型垂直/短距起降（vertical/short takeoff and landing, V/STOL）飞行器的动态过渡过程模型,综合考虑过渡走廊限制、操纵冗余及不同起降任务需求指标,研究最优过渡操纵策略。考虑V/STOL飞行器的喷射气流效应,对飞行器进行全量动力学建模。利用可达平衡集方法,建立通用过渡走廊计算框架。设计了能够在V/STOL过渡段和高速飞行间平稳过渡的操纵方式。将推力矢量飞行器的动态倾转过渡过程转化为非线性动态最优控制问题,根据不同起降任务特点建立合理的指标和约束,采用直接转换法和序列二次规划算法进行求解,得出不同任务特点下的最优操纵策略与过渡过程。采用可达平衡集计算过渡走廊的方法,不仅不受飞行器类型的限制,更简化了构造过程,具有良好的通用性与鲁棒性。以光滑过渡为目标的优化结果使得飞行员在飞行器过渡过程中的操纵量变化大幅减小,从而使得飞行员能更加专注于对飞行器运动的操纵;以距离更短为目标的优化结果则使得降落过程的飞行距离缩短了30%左右。从操纵策略出发的优化结果使得驾驶员能够更好掌握操纵关注点及边界,增加了整个动态过渡过程的安全性。     

An Anthropocentric Approach for Solving the Problem of Optimal Automatic Landing Flare

In this paper, the algorithm of automatic landing flare is designed based on the maximum coordination between the trajectories, which are convenient for the pilot to operate in manual mode and the one attained by the auto-landing system according to the algorithm of optimal flight control.     

Numerical study on flow control of ship airwake and rotor airload during helicopter shipboard landing

A numerical study on flow control of ship airwake during shipboard landing is carried out to address the effect of flow control devices on helicopter rotor airload. The in-house Reynolds Averaged Navier-Stokes (RANS) based solver Rotorcraft AeroDynamics and Aeroacoustics Solver (RADAS), with combination of momentum source approach is employed to conduct the helicopter shipboard landing simulation. The control effects of three aerodynamic modifications of ship superstructure, i.e. ramp, notch and flap, in different Wind-Over-Deck (WOD) conditions are discussed. From the steady simulation results, the effect of spatial variation of ship airwake on rotor airloads is concluded. The aerodynamic modifications reduce the strength of shedding vortex and increase rotor normal force through delaying and relieving flow separation, and therefore are beneficial to alleviate the limitation of control inputs. By contrast, the perturbation of unsteady ship airwake can cause the serious oscillation of rotor forces during shipboard landing. The unsteady simulations show that the turbulence intensity of ship airwake and oscillatory rotor airloading, represented by Root-Mean-Square (RMS) loading, can be remarkably reduced by the ramp and notch modifications, while the flap modification has adverse effect. It means that flow control devices have large potential benefits to alleviate the pilot’s workload and improve the shipboard landing safety, but they should be well designed to avoid the introduction of more vortex, which leads to increase in disturbance of flow field.     

基于自抗扰控制的共轴直升机姿态控制律设计

以某型共轴双旋翼无人直升机作为研究对象,对无人机返航进场与降落着舰阶段的抗扰动问题进行研究。针对海上着舰环境下的外界干扰与不确定性问题,选用自抗扰控制器设计共轴直升机姿态控制律,以提高共轴直升机对外部扰动的抗扰性能。针对自抗扰控制律调整参数过多问题,设计改进粒子群算法进行自动参数优化。仿真结果表明:改进粒子群算法能够显著提高参数整定速度,且精度更高;优化参数后的自抗扰控制律在强干扰环境下具备良好的动态响应与鲁棒性。     

跟踪控制状态飞行员建模研究

在跟踪飞行状态时,飞行员是按被跟踪状态量的速率信息和跟踪误差信息以一定的控制规律进行操纵的。研究了飞行员在跟踪控制状态时操纵模型的建立方法,即如何建立飞行员的控制规律。以光学助降人工着舰时飞行员下滑"m eatball"对中跟踪操纵为例,采用海军惯用的"backside"技术设计了飞行员控制规律的操纵增益参数,并辅以仿真验证。结果表明,跟踪控制状态飞行员建模的简化方法具有普遍意义。     

人—机闭环系统的驾驶员模型研究

通过对驾驶员操纵行为的分析，提出了在人－机闭环系统研究中驾驶员环节数学描述的要求，对驾驶员的传递函数模型，最优控制模型和模糊控制模型进行了分析研究，并对驾驶员模型在人－机闭环系统动态品质分析方面的应用进行了研究。