Control Design Of 2DOF Servo Cloud-platform System Based On Dual-input Dual-output System Identification

In recent years,with the rise of advanced technologies such as unmanned aerial vehicles and unmanned vehicles,the performance requirements of the servo cloud-platform system mounted on them have been continuously improved.To follow up the rapid development of technology and meet the various high-performance requirements of the servo system,the control technology of the servo cloud-platform system has important and urgent research value.At present,the PID control scheme based on the classic control theory has gradually become weak under the multi-objective control performance requirements such as fastness,stability,robustness and anti-disturbance,and the control design method based on modern control theory is likely to become a breakthrough to further meet multi-objective control performance requirements.For the servo cloud-platform system,there are many flexible modes and nonlinearities in the structure,causing multiple resonance in the system.In the two-degree-of-freedom(2DOF)servo cloud-platform system,it is easier to cause coupling between axes,which increases the complexity of the system and the difficulty of control design.As for system decoupling,there is still a lot of work to be done in both theoretical research,development and application.In order to study the control problems in the 2DOF cloud-platform system,this paper builds a cloud-platform system with 2DOF rotation positioning function,simulating the cloudplatform in industry.For the cloud-platform system,first,the various nonlinearities links that may exist in it and design algorithms to compensate nonlinearity to increase the linearity of the system was analyzed.Then,by inputting orthogonal pseudo-random signals,Hankel matrix method based on impulse response is used for system identification,and a state-space model of a dual-input dual-output system is obtained.Then the balanced truncation method to reduce the order of each identified subsystem to obtain the cloud-platform pitch axis model,azimuth axis model and inter-axis coupling model is used,which is convenient for the independent control design of each subsystem and reduces the complexity of the control design.In the control design,considering the high real-time requirement of the servo cloud-platform system,the model matching algorithm to design the feedforward decoupling controller is used to replace the advanced decoupling controller which is difficult to be practically applied in this paper.For the control design of the pitch and azimuth axis subsystems of the cloud-platform,a dual closedloop structure control system to improve the anti-disturbance performance of the system is used.While LQ controller based on the integral internal model is used in the speed loop system,comparing with the traditional PI controller.A simple lag and PD controller to prevent the controller from being too large while increasing the responding speed of the system is used in external loop.Finally,the discretization method of all the above controllers is also provided in this paper and designed controllers are verified in the actual 2DOF servo cloud-platform system.