航空遥感惯性稳定平台模糊/PID复合控制

为提高航空遥感惯性稳定平台控制系统稳定精度和扰动抑制能力,在常规PID控制的基础上设计了一种模糊控制与PID相结合的复合控制算法,分别应用于稳定平台横滚框及俯仰框系统进行实验验证。在三环控制系统位置环中将模糊控制与PID控制方法结合使用,并引入变论域思想,建立模糊/PID复合控制器,满足输出偏差变化不同时刻对PID参数整定的要求。通过模糊控制器实时调整PID参数,使系统具有良好的动、静态特性,实现多源扰动下惯性稳定平台的高稳定精度控制。分别通过仿真和静动态实验对方法进行分析和验证。实验结果表明:与常规PID控制及单纯模糊控制相比,模糊/PID复合控制器具有优越的扰动抑制能力和高稳定精度。相对传统PID控制,横滚框和俯仰框的静态均方根误差(RMS)值分别下降51%和73%、动态RMS值分别下降约20%和30%。

Fuzzy/PID compound control for inertially stabilized platform in airborne remote sensing

High precision aerial remote sensing imaging needs inertial stabilized platform (ISP) to isolate attitude variations of aircraft and the multi source disturbances. To improve the control system stabilization accuracy and disturbance rejection ability of ISP, on the basis of conventional PID control, a composite control algorithm is proposed combining both of fuzzy and PID control methods, which is respectively applied in both of the roll gimbal and pitch gimbal control systems for experiment verification. In the position loop of the three loop control system, the fuzzy and PID control methods are combined and used; in the controller design, the idea of variable universe is used to establish the fuzzy/PID compound controller. In this way, the PID parameters are adjusted in real time on the basis of the deviation change in different time given by the fuzzy controller. Thus, the good dynamic and static characteristics of the system are obtained, the high stabilization accuracy control of ISP under multi source disturbances is achieved. To verify the proposed method, simulation and static and dynamic experiments were conducted. The results show that the fuzzy/PID compound controller has excellent disturbance rejection ability and high stabilization accuracy compared with only the conventional PID and pure fuzzy controls. Compared with PID control, the static root mean square error (RMS) values of the roll gimbal and pitch gimbal decrease by 51% and 73%, respectively, while the dynamic RMS values decrease by about 20% and 30% for the two gimbals, respectively.