【作者】 许德新；

【导师】 谈振藩；

【作者基本信息】 哈尔滨工程大学 ， 导航、制导与控制， 2011， 博士

【摘要】 无人机光电载荷视轴稳定系统作为无人机的“眼睛”,对于空中侦察、监视等起着举足轻重的作用。该系统主要完成视轴稳定、目标跟踪两种功能,系统的核心与难点是视轴稳定的控制技术；作为系统的核心元件,MEMS惯性器件具有体积小、成本低、测量范围大、可靠性高和易于数字化等优点,适合用于小型无人机对载重较为敏感的场合。对于小型无人机光电载荷视轴稳定系统而言,由于空中环境的复杂性,如何有效地隔离载体的机动及外界的扰动并对光电稳定平台实现高精度、高可靠、高稳定的控制显得尤为重要。传统的控制方式建立在被控对象模型参数固定的基础上,但是在实际情况中,被控对象的参数如转动惯量、摩擦力矩等并不是保持不变的,如何采取有效的鲁棒控制方式以提高系统的视轴稳定精度及跟踪性能显得迫切需要。论文首先对无人机光电载荷视轴稳定系统进行了系统分析与数学建模。针对系统设计中考虑的主要问题进行了详细分析,包括系统控制器的软硬件方案,稳定轴数目的选取与运动分析,鲁棒控制必要性分析,MEMS陀螺去噪与温度补偿。针对MEMS陀螺精度受环境温度影响的问题,提出了一种MEMS陀螺全温度标定与补偿方法,得到优化的MEMS陀螺分段误差补偿模型并进行相应的补偿。实验结果表明,该方法能有效地解决MEMS陀螺温度补偿问题。在分析系统各环节模型不确定参数的基础上,针对三轴光电载荷视轴稳定系统的参数不确定性,提出了将定量反馈(Quantitative Feedback Theory,简称QFT)理论设计视轴稳定系统鲁棒控制器的策略。给出了无人机光电载荷视轴稳定系统的QFT模型并进行了分析,将QFT控制器与PID控制器的性能进行了比较。仿真结果表明,QFT控制器的性能要优于传统的PID控制器。采用QFT设计方法,可以解决控制系统在参数变化导致性能指标下降的情况,具有很强的鲁棒性。针对光电载荷视轴稳定系统的多输入多输出(MIMO)特点,进行了MIMO系统QFT控制方法研究,提出了一种MIMO系统QFT控制器的解耦及简化设计方法。首先将MIMO系统的控制问题转化为一系列MISO子系统,并采用近似不相关方法减少MISO子系统个数,在保证系统性能不受影响的前提下,简化了MIMO系统的QFT设计过程。推导过程与结果表明,该方法对于解决多输入多输出系统的QFT控制问题非常有效。在定量反馈理论控制器设计的基础上,提出了一种QFT/H∞组合控制器的设计方法。分析了QFT与H。的区别与关系,给出了QFT/H∞组合控制器设计的详细步骤,并且将QFT/H∞组合控制器的性能与传统QFT控制器进行了比较。仿真结果表明,该组合方法组合控制器具有更理想的性能指标,即具有更短的调节时间和更小的稳态误差。除了比传统QFT设计方法方便之外,还结合了两种控制方法的优点。论文最后进行了变参数条件下系统的闭环仿真,仿真结果表明,传统的控制器超调量、稳态误差较大,而QFT控制器可保证系统在模型参数摄动、负载扰动时具有较强的鲁棒性,且调节时间较短,超调很小,满足设计所要求的性能指标。更多还原

【Abstract】 As the eye of unmanned aerial vehicle, line-of-sight stabilization system of opto-electronic load is significant in the task of aerial scouting. It has two main functions, one of them is line-of-sight stabilizing and the other is object tracking. Focus of the system is the control technology of line-of-sight stabilization. As the key device of the system, MEMS inertial device is used for the mini-type UAV because it has a lot of advantages such as tiny volume, low cost, large measurement range, better reliability and easy to be digitalized.Because of the complex aerial environment, it is very important that how efficiently to isolate the shaking and disturbing and realize high accuracy, high reliability, high stability control to the optoelectronic stabilized platform. The traditional control method is based on the hypothesis that all of the parameters are constant, but in fact some of the parameters of controlled object will change such as the gimbal’s moment of inertia and the friction moment. It is very important to enhance the system’s stabilizing precision and tracking performance by adopting efficient robust control.Firstly, Line-of-sight stabilization system of opto-electronic load for UAV is analyzed and the mathematic model is built. The main problem for system design is analyzed, which includes the hardware and software plan for the system controller, the choosing of stabilizing axis number and scheme, the movement analysis of stabilized axis, the necessity analysis of robust control and the denoising and temperature compensation for MEMS gyro. According the problem of environment temperature influencing on MEMS gyro, A whole temperature range demarcating and compensation method is brought up, the temperature error is compensated after the optimal subsection error compensation model analysis. The test indicates that this method can solve the MEMS gyro temperature compensation problem effectively.Basing on the analysis of uncertain parameters of system model, According to the parameters uncertainty of three-axis line-of-sight stabilization system of opto-electronic load, A Quantitative Feedback Theory method is brought forward to design the robust controller for line-of-sight stabilization system. The QFT model of line-of-sight stabilization system of opto-electronic load for UAV is analyzed, the performance of QFT controller and PID controller is compared. The simulation result indicates that performance of QFT controller is better than PID controller. QFT control method has good robustness, and it can solve the problem of performance decreasing which occurs under the situation of parameters changing in the control system.According the characteristic of multi-input and multi-output for the line-of-sight stabilization system of opto-electronic load, the MIMO Quantitative Feedback Theory control method is studied, A decoupling and simplifying design method is put forward for the QFT controller of MIMO system, which firstly converts the MIMO system control problem to a series of MISO sub-system. Secondly adopts basically noninteracting method to reduce the number of MISO sub-system. Thus the QFT design for MIMO system is simplified and has no effect on the system performance. The analysis result indicates that this method is effective for solving the problem of controller design for MIMO system.Basing on the QFT controller analysis and design, a combined method is put forward which integrate the Quantitative Feedback Method and H∞control method. The difference and relationship between these two methods is analyzed and the detail procedure of integrated controller design is given, The performance result is compared between the QFT controller and the integrated controller, the simulation indicates that this integrated method has more ideal performance than QFT method, it has shorter adjust time and less steady-state error. Not only it is more convenient, but also can it combine the advantages of two kinds of different control methods.Finally the closed-loop simulation for the system is done under the condition of variant parameters, the simulation result indicates that traditional controller has larger overshoot and steady-state error, QFT controller has better robustness performance under the condition of model parameters changing and load disturbing, the more important thing is that QFT controller has shorter adjust time and less overshoot, it can fit the performance request of system effectively.更多还原