基于模型预测控制的气浮台运动控制研究

发布时间：2018-06-10 17:53

本文选题：气浮台 + 运动控制系统设计　；参考：《哈尔滨工业大学》2017年硕士论文

【摘要】：气浮台作为一种借助平面型气浮轴承在一个平面上实现几乎无摩擦运动的航天器地面模拟装置,以其成本相对较低、持续时间较长、引入扰动较小等优点在近几十年发展迅速,被广泛应用于控制灵活机械手、机械手硬件测试、自由飞行的卫星机械手系统、卫星编队飞行与临近操作、交会对接(捕捉)、低重力机构登陆装置等主题的航天任务研究中。作为这些研究的基础,首先要实现对气浮台的运动控制。本文以一种3自由度气浮台为研究对象,详细介绍气浮台运动控制系统的设计与实现过程,并尝试采用模型预测控制器进行控制。首先介绍气浮台的基本组成与关键参数。根据理论力学的相关知识,建立惯性坐标系与体坐标系,对气浮台进行动力学分析,分别推导其在两种坐标系下的动力学模型。此外,还介绍一些模型预测控制的知识。其次搭建气浮台运动控制系统并完成星载计算机上的软件架构设计,并设计实现一些基本模块。首先采用一种非线性跟踪微分器作为观测器,对其功能分别进行仿真与实验验证;其次根据喷嘴的安装位置设计并实现推力变换分配算法;然后设计信号调制模块,对需要PWM调制部分,给出一种经典的PWM占空比的计算方法,对需要PWPF调制的部分,选取合理的参数并进行仿真验证。然后进行控制器的设计工作。首先分析电磁阀开关特性的影响并设计PID控制器,实现气浮台的基本运动控制并与将要设计的模型预测控制器作对比。在此基础上尝试设计模型预测控制器。对两种控制器分别采用阶跃信号、斜坡信号、正弦信号进行仿真与试验验证,结果表明PID控制器可以满足试验需求,模型预测控制器尽管性能受到星载计算机硬件限制,仍可以用于运动控制。最后从气浮台的主要应用领域抽象出自由飞行与定点悬浮、矩形轨迹跟踪、圆轨迹跟踪3中典型运动形式,并用对两种控制器对以上运动控制的有效性进行了数值仿真与试验验证,结果表明两种控制器都实现了对上述典型运动的控制。由于模型预测控制器性能受限,PID的控制效果显得相对较好,但是前者也显示出控制过程更平稳直接、抗干扰能力更好、控制量始终在执行器的输出范围内(主动显式的处理约束)、使得执行器不至于长时间陷入饱和状态、控制量输出较小节省燃料等优点。
[Abstract]:Air floatation platform, as a kind of spacecraft ground simulation device which realizes almost frictionless motion on a plane by means of plane air bearing, has developed rapidly in recent decades because of its advantages of relatively low cost, longer duration, less disturbance and so on. It is widely used to control flexible manipulator, hardware testing of manipulator, free-flying satellite manipulator system, satellite formation flight and proximity operation, rendezvous and docking (capture and docking), low-gravity mechanism landing device and other topics of space mission research. As the basis of these studies, the first step is to realize the motion control of the air-floating platform. In this paper, the design and implementation of the motion control system of a three degree of freedom air floatation platform are introduced in detail, and the model predictive controller is used to control the system. First, the basic composition and key parameters of the air flotation platform are introduced. According to the relevant knowledge of theoretical mechanics, the inertial coordinate system and the body coordinate system are established, and the dynamic models of the air floatation platform are derived respectively. In addition, some knowledge of model predictive control is also introduced. Secondly, the motion control system of the air floatation platform is built and the software architecture on the spaceborne computer is designed, and some basic modules are designed and implemented. Firstly, a nonlinear tracking differentiator is used as observer, and its function is simulated and verified by experiments. Secondly, according to the installation position of the nozzle, the thrust transform assignment algorithm is designed and realized. Then, the signal modulation module is designed. A classical PWM duty cycle calculation method is presented for the PWM modulation part. The reasonable parameters are selected and verified by simulation. Then the controller is designed. Firstly, the influence of solenoid valve switching characteristics is analyzed and pid controller is designed to realize the basic motion control of the air floatation platform and compare it with the model predictive controller to be designed. On this basis, we try to design a model predictive controller. The two controllers are simulated and verified by step signal, ramp signal and sinusoidal signal respectively. The results show that the pid controller can meet the test demand, and the performance of the model predictive controller is limited by the hardware of spaceborne computer. Can still be used for motion control. Finally, from the main application fields of the air floatation platform, the typical motion forms of free flight and fixed-point suspension, rectangular trajectory tracking and circular trajectory tracking are abstracted. The effectiveness of the two controllers for the above motion control is numerically simulated and tested, and the results show that the two controllers can control the typical motion mentioned above. Because the performance of the model predictive controller is limited, the control effect of pid is relatively good, but the former also shows that the control process is more stable and direct, and the anti-jamming ability is better. The control quantity is always within the output range of the actuator (active and explicit processing constraints, so that the actuator does not fall into saturation state for a long time, the output of the control quantity is small to save fuel and so on.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：V416.8;TP273

【参考文献】