基于输出重定义的非最小相位系统轨迹跟踪控制

发布时间：2018-05-29 13:50

本文选题：非最小相位系统 + 输出轨迹跟踪　；参考：《浙江大学》2017年硕士论文

【摘要】：非最小相位系统常见于工程应用场合,如应用于航天和生产制造领域的柔性机械臂,即是一种典型的非最小相位系统。由于不稳定零动态的影响,非最小相位系统不存在因果稳定的逆,因此常规反馈控制不能实现精确轨迹跟踪。非因果稳定逆控制是实现非最小相位系统轨迹准确跟踪的本质方法。但由于其理论上需要有无限时长的预作用,不能直接应用于实际系统操作;而且稳定逆控制算法的有效性仅局限于双曲非最小相位系统。综合考虑实际应用及跟踪精度要求,本文针对双曲和非双曲非最小相位系统,采用近似线性化方法,提出轨迹重定义的设计思路,以基函数形式重新定义期望输出轨迹,实现被控系统在有限时间区间上的因果逆轨迹跟踪,且系统实际输出与理想期望轨迹的误差满足给定的跟踪精度要求。本文取得的具体成果如下:(1)为解决非因果稳定逆无限时长前驱动不适用于实际操作的问题,本文提出在有限时长区间内重定义输出轨迹取代无限时长前驱动过程,实现了被控系统由实际起始状态到稳定逆理想初始状态的过渡。然后结合非因果稳定逆的控制方法实现了非最小相位系统的准确轨迹跟踪控制。重定义的输出轨迹采用指数基函数的形式,轨迹重定义过程使用一种优化搜索算法寻找到最优轨迹。在一个单连杆的柔性机械臂系统上进行了新控制方法的仿真。结果说明了控制方法的有效性。(2)为设计适用于广义非最小相位系统的轨迹跟踪控制算法,同时提高轨迹跟踪精度,本文提出了因果逆分段轨迹跟踪控制方法。以指数基函数形式重新定义整段输出轨迹,在尽可能近似期望输出轨迹的条件下,设计控制器抵消掉非最小相位系统中不稳定零点的影响,抑制发散。将输出轨迹划分为有限段逐段跟踪,同时保证输入信号、整段输出轨迹及系统内动态轨迹的连续性。将分段轨迹跟踪控制算法应用于实验室悬挂式单连杆柔性臂系统,仿真结果有效地证明了新方法能提高轨迹跟踪精度。(3)实验验证,将所提出的分段轨迹跟踪因果逆控制方法应用于一个典型的非最小相位系统——柔性臂,通过最小二乘法辨识实际柔性臂系统模型,离线计算其控制输入信号,反向输入到实验系统,观察记录实际输出信号,与期望输出轨迹进行对比,对结果进行分析。实验结果很好地证明了所设计控制方案的可行性。
[Abstract]:Non-minimum phase systems are commonly used in engineering applications, such as flexible manipulators used in aerospace and manufacturing fields, that is, a typical non-minimum phase system. Due to the effect of unstable zero dynamics, there is no causal stability inverse in the non-minimum phase system, so the conventional feedback control can not achieve accurate trajectory tracking. Non-causal stable inverse control is an essential method for accurate trajectory tracking of non-minimum phase systems. However, it can not be directly applied to practical system operation because of the need of pre-action with unlimited time in theory, and the effectiveness of the stability inverse control algorithm is limited to hyperbolic non-minimum phase systems. Considering the practical application and the requirement of tracking precision, this paper proposes a new design idea of trajectory redefinition for hyperbolic and non-hyperbolic non-minimum phase systems by using approximate linearization method. The desired output trajectory is redefined in the form of basis function. The causal inverse trajectory tracking of the controlled system in a finite time interval is realized, and the error between the actual output of the system and the ideal desired trajectory meets the given tracking accuracy requirements. The concrete results obtained in this paper are as follows: 1) in order to solve the problem that the non-causal stable inverse infinite time front drive is not suitable for practical operation, this paper proposes to redefine the output trajectory in the finite time interval to replace the infinite time front drive process. The transition from the actual initial state to the stable inverse ideal initial state of the controlled system is realized. Then the accurate trajectory tracking control of the non-minimum phase system is realized by using the non-causal stable inverse control method. The output trajectory is redefined in the form of exponential basis function. The trajectory redefinition process uses an optimal search algorithm to find the optimal trajectory. The simulation of the new control method is carried out on a single link flexible manipulator system. The results show the effectiveness of the control method. In order to design a trajectory tracking control algorithm for generalized non-minimum phase systems and improve the tracking accuracy, a causal inverse piecewise trajectory tracking control method is proposed in this paper. The whole output trajectory is redefined in the form of exponential basis function. Under the condition of approximating the desired output trajectory as much as possible, a controller is designed to counteract the influence of unstable zeros in the non-minimum phase system and to suppress divergence. The output trajectory is divided into finite segment by segment tracking, and the continuity of input signal, whole output track and dynamic trajectory in the system is ensured at the same time. The piecewise trajectory tracking control algorithm is applied to the single link flexible arm system in the laboratory. The simulation results show that the new method can improve the trajectory tracking accuracy. The proposed piecewise trajectory tracking causality inverse control method is applied to a typical non-minimum phase system-flexible arm. The actual flexible arm model is identified by the least square method, and the control input signal is calculated offline. The actual output signal is observed and recorded, and compared with the desired output trajectory, and the results are analyzed. The experimental results show the feasibility of the proposed control scheme.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP241

【参考文献】