| The moving base optoelectronic stabilized platform is a kind of integrated optoelectronic system integrating light,machinery and electricity,which can carry different loads to achieve specific functions.It is widely used in various fields of military and civil.In the process of tracking and detection,it is affected by internal and external disturbances such as carrier motion,mechanical vibration,various interfering torques and measurement noise,which will lead to the decrease of measurement accuracy.Therefore,it is necessary to adopt appropriate structure and control system design to compensate the interference of these factors to the optic axis pointing of the photoelectric stabilized platform.The stabilization technology of optic axis is a key technology to develop the photoelectric stabilization platform with moving base.At present,digital stabilization method is widely used to realize the self-stabilization function of visual axis of photoelectric stabilized platform with moving base.There are mainly two kinds of digital self-stabilization methods: the method of axis of view stabilization based on inertial data geographic information and the method of axis of view stabilization based on rate gyro inertial information.In this paper,several key problems of optic axis stability of photoelectric stabilized platform,including the two optic axis stabilization methods,system modeling and controller design,are studied.The work and innovation of this paper are as follows:Firstly,the structure classification of the stabilized platform with moving base and the coordinate system used to deduce two kinds of axial-of-view stability principles are introduced.This paper introduces the concrete derivation process of two kinds of axialof-view stability principle and selects the installation mode of rate gyro according to the derivation formula.The factors affecting the stability accuracy of the electro-optical stabilized platform are analyzed,including disturbance distribution characteristics of the carrier,mechanical characteristics and sensor characteristics.It provides a research basis for improving the stability accuracy of optic axis of photoelectric stabilized platform.Secondly,permanent magnet synchronous motor is selected as the driving motor of photoelectric stabilization platform,and the mathematical model and vector control strategy of the motor are introduced.According to the classical frequency domain method,each loop controller of the electro-optical stabilized platform servo loop is designed.Stribeck friction torque model is established and introduced into the velocity loop as interference torque.Linear active disturbance rejection control,which is easy to implement in engineering,is applied to speed loop to suppress the influence of internal and external disturbances.A prediction algorithm is proposed to solve the problem of lag effect on observer performance caused by speed sensor.The simulation experiments of linear active disturbance rejection controller,linear active disturbance rejection controller and traditional PI controller are carried out.The experimental results show that under the same closed-loop bandwidth,the step response of linear active disturbance rejection control system based on prediction and linear active disturbance rejection control system has no overshoot.Compared with the linear active disturbance rejection control system,the step response stability time of linear active disturbance rejection control system based on prediction is reduced by 10% and the RMS value of velocity fluctuation is reduced by 10.36%.Compared with the PI control system,the step response stability time of linear active disturbance rejection control system is reduced by 60.87% and the RMS value of velocity fluctuation is reduced by64.73%.When the amplitude is 1°/s and the frequency is 2Hz,the isolation degree of the predictive linear active disturbance rejection control system is improved by 15.83%compared with the linear active disturbance rejection control system,and by 44.86%compared with the PI control system.Then,several key problems of optic axis stability of opto-electronic stabilized platform are emphatically studied.The first problem is how to design the tracking strategy for the moving target according to the given input when tracking the moving target using the axis of sight stabilization method based on inertial derivatives according to geographic information.Therefore,this paper classifies the working conditions,improves the original axis of sight stabilization algorithm and gives the corresponding stabilization strategy.Then,one of the working conditions is taken as an example for simulation verification,and the curve of sensor error and pitch error is fitted,which provides a basis for the selection of inertial navigation system and encoder.The second problem is that the line of sight drift will occur when the method of axis of view stabilization based on the inertial information of rate gyro is used.The compensation algorithm is verified by simulation.Monte Carlo method is used to analyze the error factors affecting the compensation accuracy and the corresponding curve is fitted.The third is the study of the gis data filtering algorithm,the tracking differentiator is applied to the filtering,and compared with the second order Butterworth filter and Kalman filter simulation verification.The results show that the tracking error RMS value of the feedforward compensation system using tracking differentiator filter decreases by 69.23%compared with the system using Kalman filter and 38.46% compared with the system using second-order Butterworth filter.Finally,software development is carried out on the established hardware system,and the specific design and development process are given.The above algorithms are verified by experiments.The experimental results of the speed loop controller are as follows: compared with the PI control system,the step response stability time of the linear active disturbance rejection control system based on the forecast is reduced by32.54%,the overshoot is reduced by 88.01%,and the RMS value of the velocity fluctuation is reduced by 33.53%.When the amplitude is 1°/s and the frequency is 2Hz,the isolation degree of the predictive linear active disturbance rejection control system is 53.81% higher than that of the PI control system.Experimental verification results of logarithmic tracking of one of the working conditions--swing sweep mode show that the maximum fluctuation of pitch Angle error is 0.12° in the geographic coordinate system after using the compensation algorithm,which is much smaller than 10.1°without using the compensation algorithm.The experimental verification results of the feedforward compensation algorithm for line-of-sight drift are as follows: in the test time of 30 min,when the pitching axis of the photoelectric stabilized platform is oriented towards the north and south,the pitching Angle drift after the compensation algorithm is 0.231°,which is far less than 6.005° without the compensation algorithm.When the pitching axis of the electro-optical stabilized platform is oriented toward the north and south,the pitching Angle drift is-0.385° after using the compensation algorithm,which is far less than-5.372° without using the compensation algorithm.Experimental verification of gis data filtering algorithm: the tracking error RMS value of feedforward compensation system filtered by tracking differentiator is reduced by 42.59% compared with that of kalman filter,and 47.46% compared with that of second-order Butterworth filter.The experimental results prove the effectiveness of the proposed algorithm,which has high practical value for improving the stability accuracy of visual axis of moving base photoelectric platform. |
