Research On The Reentry Trajectory Optimization Method Of Space Maneuver Vehicle

Because of the reusability、load capacity and flexibility,Space Maneuver Vehicle(SMV)has been an important research content of aerospace field.Trajectory optimization is a major part of the overall design of SMV,which affects the aerodynamic layout,guidance control,and other subsystems.In addition,reentry is the most representative part of the return process of an aircraft.Therefore,it is significant to study on the reentry trajectory optimization of space maneuver vehicle.In this thesis,the following work is completed:(1)Establishing the dynamic model of space maneuver vehicle,aiming at shortest time.In this model,the Angle of Attack and the Bank Angle are chosen as control variables and eight dimensional state variables such as speed and altitude are selected.Moreover,the boundary,terminal and path constraint conditions are set up to meet the mission requirements of the flight safety.(2)Applying dynamic neural network method,which combined neural network with indirect method,to solve the reentry trajectory optimization problem of SMV.Compared with the direct method,indirect method will obtain a higher accuracy solution,while adding neural network can avoid the problem of traditional indirect method that it has to guess the covariates’ initial value,and simplify the calculation process of indirect method.(3)Using genetic algorithm to optimize the initial connection weights of dynamic neural network.With the favorable global search ability of genetic algorithm and the self-learning ability of neural network,a more accurate global optimal solution will be obtained.The simulation results show that the method adopted in this thesis can achieve the optimal solution of reentry trajectory of SMV,and has higher accuracy and robustness compared with the traditional methods.