任务空间内空间机器人鲁棒智能控制器设计

研究了基于神经网络的自由漂浮空间机器人在任务空间内的轨迹跟踪问题。首先利用RBF神经网络来逼近自由漂浮空间机器人高度非线性的动力学模型，然后设计了鲁棒控制器对逼近误差和外部干扰进行抑制。利用Lyapunov直接方法建立的新的神经网络参数和连接权值的在线学习算法，以及利用耗散理论设计的鲁棒控制器保证了系统的稳定性，并能够使系统L2增益小于给定的指标。利用该控制器对平面二连杆空间机器人进行了仿真研究，表明该智能控制方案是有效的。     

Stewart平台动力学建模及鲁棒主动隔振控制

未来复杂航天器低频模态密集,其敏感载荷要求很高的指向精度和稳定度,只对航天器本体姿态控制很难满足要求.本文采用Stewart超静平台对敏感载荷进行6自由度主动隔振,建立了非线性动力学模型,并根据线性模型设计了多变量鲁棒控制器,采用DK迭代算法求解.频域分析可得Stewart平台对3～800Hz的扰动主动隔振大于25dB,仿真证明Stewart平台对10Hz谐波扰动隔振性能优于40dB,对白噪声随机扰动隔振性能优于30dB,有效抑制了微小扰动,起到了6自由度超静隔振平台作用.     

弹性体运载火箭建模及控制器设计

考虑运载火箭飞行过程中的弹性振动特性,对运载火箭进行全量动力学建模以及控制器设计。首先将运载火箭考虑为一维梁模型并建立全量动力学模型,分别引入自适应滤波姿态控制算法和H_2范数鲁棒增益调度控制算法设计控制器。仿真结果表明:在考虑舵机非线性情况下两种控制方法均能够满足精度要求,在具有较大外部扰动情况下,鲁棒增益调度控制算法相较于自适应滤波算法具有较强的鲁棒性。     

Robust adaptive fault-tolerant control of a tandem coaxial ducted fan aircraft with actuator saturation

This paper is concerned with the robust adaptive fault-tolerant control of a tandem coaxial ducted fan aircraft under system uncertainty, mismatched disturbance, and actuator saturation. For the proposed aircraft, comprehensive controllability analysis is performed to evaluate the controllability of each state as well as the margin to reject mismatched disturbance without any knowledge of the controller. Mismatched disturbance attenuation is ensured through a structured H-infinity controller tuned by a non-smooth optimization algorithm. Embedded with the H-infinity controller, an adaptive control law is proposed in order to mitigate matched system uncertainty and actuator fault. Input saturation is also considered by the modified reference model. Numerical simulation of the novel ducted fan aircraft is provided to illustrate the effectiveness of the proposed method. The simulation results reveal that the proposed adaptive controller achieves better transient response and more robust performance than classic Model Reference Adaptive Control (MRAC) method, even with serious actuator saturation.     

固体火箭的鲁棒自适应耗尽关机制导方法研究

针对固体火箭的耗尽关机制导问题,提出了一种基于闭路制导和零射程线相结合的自适应耗尽关机制导方法,该方法首先利用闭路制导技术,使待增速度快速收敛;随后基于零射程线理论,设计了对关机时间变化不敏感的零射程线能量耗散管理导引律和末速精调导引律,提高了耗尽关机方法在推力、秒耗、关机时间随机摄动条件下的鲁棒性和自适应性以及制导精度,从而拓展了耗尽关机显示制导方法的工程化应用.     

迎角受限的高超声速飞行器固定时间鲁棒控制

针对高超声速飞行器动力学模型强耦合、非对称时变迎角限制、气动参数高度不确定以及跟踪误差收敛速率要求高等问题,设计了一种考虑非对称时变迎角限制的高超声速飞行器固定时间非奇异切换控制策略。为了解决非对称时变迎角限制问题,首先限制迎角虚拟控制器的幅值并设计固定时间误差补偿系统补偿迎角虚拟控制器饱和带来的不利影响,然后设计了一种新的光滑切换的非对称时变障碍函数限制迎角跟踪误差,从而使迎角满足非对称时变限制。光滑切换技术以及固定时间收敛技术也应用于其他虚拟控制律和实际控制律的设计中,以避免奇异值问题并且保证闭环系统的固定时间稳定。此外,设计了一种固定时间稳定的鲁棒补偿器用以补偿系统不确定性带来的不利影响。严格的数学推导证明了本文方法的正确性,仿真结果验证了本文方法的有效性和优越性。     

Aeroelastic two-level optimization for preliminary design of wing structures considering robust constraints

An aeroelastic two-level optimization methodology for preliminary design of wing struc- tures is presented, in which the parameters for structural layout and sizes are taken as design vari- ables in the first-level optimization, and robust constraints in conjunction with conventional aeroelastic constraints are considered in the second-level optimization. A low-order panel method is used for aerodynamic analysis in the first-level optimization, and a high-order panel method is employed in the second-level optimization. It is concluded that the design of the abovementioned structural parameters of a wing can be improved using the present method with high efficiency. An improvement is seen in aeroelastic performance of the wing obtained with the present method when compared to the initial wing. Since these optimized structures are obtained after consideration of aerodynamic and structural uncertainties, they are well suited to encounter these uncertainties when they occur in reality.     

频率域鲁棒性分析方法及其在飞控中的应用

主要研究了频率域鲁棒稳定性分析的新方法及其在收音机横侧控制系统中的应用。为了提高计算效率，基于Ｎｙｑｕｉｓｔ判据，提出了对数和鲁棒稳定性的新判据，为了简化多变量频率域鲁棒性分析和设计方法，进一步提出了递Ｎｙｑｕｉｓｔ阵对占优势判别式，使计算量大大减少，所提出的鲁棒性新判据在飞机横侧向控制中的应用表明，该方法可以得到讼人满意的结果。     

航天器姿态预设性能控制方法综述

针对存在不确定惯量矩阵及其它未知不确定性(如执行器故障、航天器结构发生突变等)下的航天器姿态控制问题，综述了预设性能控制及其应用的研究进展。首先阐释了预设性能控制方法的基本内涵及其关键步骤；然后分析总结了现有航天器姿态控制以及预设性能控制研究的基本概况与发展趋势；最后面向未来复杂化、智能化的空间任务对航天器控制系统提出的新需求，提出航天器预设性能控制值得研究的问题和方向。     

Novel finite-time attitude control of postcapture spacecraft with input faults and quantization

This paper investigates a novel finite-time attitude control method for the postcapture spacecraft with an unknown captured space target in the presence of input faults and quantization. First, a quasi fixed-time convergent performance function is developed to quantitatively characterize the attitude tracking performance. Then, a backstepping prescribed performance attitude controller is devised via using integral barrier Lyapunov function. Compared with the existing works, the fractional state feedback and discontinuous controller is directly avoided to achieve the fixed-time convergence rate. Namely, the proposed fixed-time controller is easily achieved online. Finally, two groups of illustrative examples are organized to validate the effectiveness and robustness of the proposed control method.