| With the vigorous development of aerospace industry,aviation remote sensing stabilization platform is widely used in navigation,positioning,aerial photography and other fields.In the process of aerial photography and other tasks,the existence of various interferences has a huge impact on the accuracy and stability of the boresight of the aerial remote sensing stabilization platform.Therefore,this paper takes the aerial remote sensing stabilization platform as the research object,analyzes and studies the unknown nonlinear interference,coupling interference and other problems that appear in the process of system movement,and optimizes the control algorithm of the aerial remote sensing stabilization platform.First,establish the mathematical model of the aerial remote sensing stable platform.A single-axis system model is established based on the classic DC brushless torque motor model.In the coordinate system of the stable platform,the system attitude is calculated according to the direction of the platform’s sight axis,and the mathematical model of the pitch,roll,and yaw rotation axis frame is derived.Taking the mutual moment of each axis as the connection,the mathematical model of the aerial remote sensing stable platform system is derived.Model and analyze the main non-linear interference of the platform.Secondly,the control algorithm to suppress nonlinear interference is studied.Analyze the unknown interference that occurs during the operation of the aerial remote sensing stabilization platform,ignore the minor interference items,and clarify that the main interference items are friction interference and model uncertainties.In order to reduce the influence of friction interference and model uncertainty items on the system,the RBF neural network is used to approximate the model uncertainty items,the minimum parameter learning method is used to estimate the weight,and the parameter estimation adaptive rate is designed to replace the neural network learning algorithm.The Stribeck friction model is used as the friction interference item to compensate the mathematical model of the system,and combined with the sliding mode variable structure control algorithm to improve the robustness of the system.Then,the decoupling control algorithm of aerial remote sensing stabilization platform is researched.Through the analysis of the mathematical model of the system,the stable platform system has coupling torque disturbance during operation.Therefore,a nonlinear cross-feedback decoupling scheme is designed to compensate the coupling term,and combined with the RBF neural network sliding mode controller to reduce unknown nonlinear disturbances,Improve system robustness.Finally,the decoupling control algorithm of the aerial remote sensing stabilization platform is optimized.Aiming at the problems of excessive initial error and long system convergence time found in the simulation experiment,the cerebellar neural network algorithm is used to optimize the system decoupling control algorithm,and the composite algorithm based on credit allocation and optimization smoothness reduces the output discretization.And the impact of over-learning on the system. |
