Research On Model Identification And Robust Control Technologies For Small-Scale Unmanned Helicopters

The small-scale unmanned helicopter has flight characteristics of helicopters,such as vertical takeoff and landing,hovering,cruising,low altitude and aggressive flight,and so on.Further more,it has the advantages of small size,lightweight,and high flexibility.It is widely used in both military and civil fields.However,the small-scale unmanned helicopter is a typical strongly coupled,unstable and time-varying multivariable nonlinear system,where autonomous flight control technology,is a key technology,and the system modeling and control system design is very difficulty therefore,the design and development of an autonomous flight control system is of great theoretical significance and practical utility.Focusing on single-rotor helicopter with a tail rotor,this paper studies modeling and controller design issues of small unmanned helicopter,including the establishment of the mechanism of the helicopter model,the study on the main channel model and the system model identification method considering coupling,the design of the robust controller based on identification model,and the associated flight test validation.Two licensing patents are obtained.The main contents are as follows:Using mechanism-modeling approach,six-degree freedom nonlinear flight mechanical model of small unmanned helicopter is set up;the main rotor and the mechanical properties of small wing servo waving motion characteristics significantly affecting the helicopter are analyzed,the complexity of waving models is reasonably reduced,four-degree freedom waving equation of the main rotor system are obtained.The helicopter model at the hovering point is linearized using the small perturbation linearization.The linear model expression is given.The various parameters are analyzed according to the expression of parameters to determine the scope of changes in the value of all parameters and some parameters,creating good conditions for the model identification provided.At the balance point with obvious linear characteristic and weaker channel coupling,the linear equation of helicopters are identified according to the frequency domain characteristics of the main channel,the model parameter uncertainty range are analyzed,Uncertainty range of the state equation parameters are identified.According to the insensitivity of the screening model parameters are not sensitive parameters,is proposed to implement a minimum order of model identification process variant,obtained main channel model.Experimental results show that identification results of the state equation model and variable-order model are capable of reasonably reflecting the main channel characteristics.Range of the model uncertainty can meet the requirements of practical system,making the design of the controller easier.For flight channel coupling significant problem,the model identification methods of unmanned helicopter considering the coupling are studied.A method of applying complex correlation function to the screening and identification of effective data in the frequency domain is proposed,which greatly increases the effective range of identification compared with the partial correlation function.The characteristics of the horizontal and vertical channel angular dynamic coupling model and heading angle channel dynamic coupling model are analyzed.The identification process of the state equation parameters considering the coupling is given.The model is validated through experiments.Results show that the error between the model output of the state equation and the measured output considering the coupling case is significantly improved.To solve the problems of the small unmanned helicopter that the system matrix is sparse and LMI equation is difficult to solve the parameters,the nonlinear delay model is used,and the H? robust controller is designed.By using the suitable Lyapunov function and proof procedure,the scopes of variables are expanded in order to make sure that the results of controller designed have a solution.A robust controller is designed for the helicopter's vertical and horizontal angle dynamic coupling model obtained from identification.Simulation results show that the H? robust controller enables the helicopter systems stable under a certain range of the controller disturbance,outside disturbances,delay and nonlinear not-modeled terms,and the output error converges quickly,thus improving the helicopter system's ability of anti-uncertain parameters and disturbances.Experiments of model identification and autonomous flight control are performed using the test platform of the small unmanned helicopter developed by the project team.Methods of the collection and pretreatment of the flight test data are studied.To solve the missing and outliers questions during the transmission of the flight data to the ground station,a time-domain data preprocessing method based on the characteristic model is designed.The correlation function of the input and output data are used to verify the validity of the pretreatment method.The PI attitude controller is designed based on the obtained helicopter main channel model.Automatic control experiment of real hovering is performed.The results indicate good control performance,especially good stability of the yaw channel easily to be disturbed.