The Application Of Nonlinear Model Predictive Control Method In The Gliding Trajectory Control

The gliding flight of projectile is an effective range-extended technology at present. In order to satisfy tactical guideline of 'large range, big terminal angle, high accuracy', the trajectory of the guided projectile with gliding flight is needed to be optimized. Because the gliding control initial point dispersion is larger and the gliding flight process is a nonlinear, time-varying, constrained and effected by random outside disturbances control process, the flight control system is also needed to be designed well. For the requirements of glide range-extended technology, the guided projectile trajectory optimization and nonlinear model predictive control system are designed and researched in the dissertation. The main contents are depicted as follows:1) According to the aerodynamic and motion characteristics of the canard configuration guided projectile, the six degree of freedom flight dynamic model is established. Based on instantaneous balance assumption, some reductions are applied to the six degree of freedom flight dynamic equations, and the longitudinal particle trajectory equation of the gliding flight projectile is proposed. This longitudinal particle trajectory equation is the numerical model for studying the guided projectile optimal gliding flight.2) Together with the guided projectile longitudinal particle trajectory equation, constraint conditions and the maximal gliding range objective function, the maximal gliding range trajectory optimizing model is established. The optimal algorithm including direct parametric method and sequential quadratic programming is applied to solve the guided projectile optimal trajectory. The gliding trajectory characteristics are obtained compared with the ordinal trajectory. Taking the gliding trajectory that is cosindering control ability as the concept trajectory for the gliding controller.3) To design guided projectile gliding controller, a nonlinear model predictive control method with an analytical control law (ANMPC) is proposed. For simplify designing, the guided projectile ANMPC gliding controller is divided into a guidance loop and an attitude control loop. The ANMPC guidance loop converts the mass point altitude and side deflection commands to angel-of-attack and side slip commands. The attitude control loop includes an ANMPC attitude controller and an ANMPC rolling stabilizing controller. The ANMPC attitude controller converts angel-of-attack and side slip commands to elevator and rudder fin deflections, and the ANMPC rolling stabilizing controller converts roll angle command to aileron fin deflection. In order to analyze the design parameters i.e. control order and predictive horizon affecting the ANMPC controller perfonnance, the ANMPC control loop is demonstrated through numerical simulation. And then, the guided projectile ANMPC gliding controller is demonstrated to track the concept gliding trajectory. The simulation results show that the ANMPC gliding controller possesses good control performance and the guided projectile achieves extending range.4) To design guided projectile gliding controller, another nonlinear model predictive control method basis on sequential quadratic programming (SNMPC) is proposed. There are some kinds of SNMPC methods, including classical SNMPC, multi-rate SNMPC and real-time SNMPC, that may be used to design the gliding controller. In order to finding a kind of SNMPC method which is more appropriate for the guided projectile fast dynamic characteristics, the introductions, simulations and analyses of the SNMPC methods and the comparisons between them are carried our in detail. The control performance of the three kinds SNMPC method is almost the same, but the control law of the classical SNMPC is solved slowly, and the control law of the multi-rate and real-time SNMPC method is solved most fast. And then the classical and real-time SNMPC methods are improved. The improved SNMPC methods decrease the on-line optimal variable numbers and make use of the useful information during the optimization progress, so that they could solve control laws faster. The improved real-time SNMPC is most appropriate for the guided projectile.5) A new integrated nonlinear model predictive control (INMPC) method is developed. INMPC method, which is obtained by appropriately synthesizing SNMPC and ANMPC methods, combines the advantages of them. SNMPC method could explicitly handle constrained control problems. ANMPC method could provide control law immediately. INMPC method avoids the disadvantages that the control delay of SNMPC method and ANMPC method lack of considering constraints, so that it is adapt to guided projectile fast dynamic characteristics and deals with constraints. An INMPC gliding controller is designed for the guided projectile and numerical simulations are demonstrated. The simulation results indicate that the INMPC gliding controller is able to meet the guided projectile nonlinear, constrained and fast time-varying characteristics, and the INMPC gliding controller possesses robustness.