Electro-optical tracking system are widely used in laser communication,target tracking,navigation and position system,and other fields.In order to complete the acquisition and tracking of the target,the composite axis structure containing a gimbal servo system and precision tracking system is usually adopted.This paper focuses on the optimized design of the control scheme in the electro-optical tracking system to achieve superior performance.The composite axis structure usually faces two kinds of control tasks.One is the set-point control based on the alarm signal.Through the design of positioning control,it is expected to achieve fast and stable positioning of the gimbal during the acquisition process.The other type is the tracking loop design based on the tracking error signal,which is used to improve the tracking accuracy and disturbance suppression ability of the precision tracking system.Firstly,this paper describes the composite axis structure that is commonly used in electro-optical tracking systems,and analyzes the significance of the composite axis structure.The state-space model of the permanent magnet synchronous motor in the gimbal servo system is introduced and deduced,as well as the traditional driving method of the permanent magnet synchronous motor.Secondly,the dynamic models of the controlled object in the gimbal servo system and the precision tracking platform are established.Finally,the problems to be solved by the electro-optical tracking control system are summarized.For the two types of control scenarios faced by the electrooptical tracking system,this paper designs a fast and smooth set-point servo system,and the tracking control algorithms with high precision and high disturbance suppression ability.In order to improve the fast response performance of the current loop in the gimbal servo system,two overmodulation strategies based on predictive current control are proposed in this paper.Firstly,a solver-based continuous control set model predictive control scheme is proposed.The control objective is abstracted as a quadratic programming problem,and the modulation voltage is solved by the active set method.Another strategy is to inject the common-mode-saturation signal into the modulation voltage obtained by deadbeat control,and demonstrate the optimal performance of this scheme from the perspective of geometric projection.Finally,in addition to proving the optimality and equivalence of these two overmodulation strategies from the perspective of theoretical derivation,sufficient simulation analysis and experimental results verify the correctness of the theoretical analysis.Meanwhile,the method proposed in this paper shows faster performance compared with the traditional current control method.To meet the fast and smooth requirements of the set-point control,a proximate time optimal positioning servomechanism based on the transition process is proposed in the set-point control of the gimbal servo system.After analyzing the problems existing in the proximate time optimal positioning servomechanism,a switching control strategy based on the transition process is proposed.For a large-scaled positioning reference signal,the proximate time optimal positioning servomechanism is directly adopted.While for a small-scaled positioning reference signal,a transition process is added,so that the step reference signal becomes a gradually changing rising signal.The transition process is derived from the optimal control theory,using the amplitude of the positioning reference signal and the limit capability of the actuator to quantify the step signal,so allows the closed-loop system to track the quantized curve.Finally,the largescaled and small-scaled reference signals are pointed in the electro-optical pod systems many times,and the conclusion that the switching control strategy is effective,fast and stable is obtained.For the requirements of high precision and high disturbance suppression ability of the precision tracking system,an error tracking method based on the type-III control loop is proposed.For any PID controller,by integrating the additional modules proposed in this paper to form a type-III control loop,the low-frequency error suppression and disturbance suppression capabilities of the system are improved,and the performance improvement frequency band of the type-III loop compared with the type-I PID loop is given.For the undetermined parameters of the type-III control loop,the multi-objective non-dominated sorting genetic algorithm is used to determine the globally optimal control parameters.The control objectives include the closed-loop bandwidth,error suppression ratio and conditional stability interval.After the proposed algorithm is verified on the precision tracking platform,the performance improvement of the frequency bands for error suppression and disturbance suppression fully verifies the theoretical results.For the fast reference input signal tracking,such as the ramp and acceleration signal,the proposed type-III control loop shows better tracking performance.Due to the low sampling frequency of the sensors and the inherent characteristics of the digital systems,the delay characteristics of the system are inevitable.Aiming at solving the parameter tuning problem when considering the delay element in the typeIII control loop,a parameter tuning method based on the amplitude optimal criterion is proposed.From the perspective of the frequency domain,this method makes the amplitude of the closed-loop frequency characteristic close to one in the widest possible frequency band,and the eighth-order Taylor expansion is adopted to approximate the delay element,then the analytic relations of the specific parameters are determined in the controller.Finally,the method in this paper has a smaller steady-state error and better disturbance suppression characteristics compared with the traditional parameter tuning method on the electro-optical pod experimental platform. |