Study On Endo-atmospheric Ascent Trajectory Rapid Optimization For Launch Vehicles

High speed, automation, low cost, high reliability and flexibility are the aims of advanced vehicles from the beginning. Traditional atmospheric ascent section of launch vehicles adopts open-loop guidance scheme, it is obvious that it cannot fulfil the needs of developing the advanced vehicles, due to the problems of high cost, long consumed time, inability of processing urgent launch tasks, low mission adaptability, poor autonomy and anti-jamming ability, and low accuracy et al. Adopting the rapid trajectory optimization technology to implement trajectory on-line planning and closed-loop guidance can solve these problems caused by open-loop guidance, meanwhile it can be utilized in rapid nominal trajectory design and the overall optimization design. It is one of the most promising schemes, which also bears practical engineering values, for atmospheric guidance of vehicles. Based on the atmospheric ascent section of vehicles, this paper studies various rapid trajectory optimization technologies and closed-loop guidance methods. The mainly study contents are listed as follows:A rapid trajectory optimization based on modified indirect method is studied. Improvements are made for enhancing solving performance of the indirect algorithm. Firstly, the first-order necessary conditions based on the optimal control transfer invert the endo-atmosphere vehicle risen trajectory optimization problem by considering path constraint into Hamiltonian two-point boundary value problems(HTPBVP), and solve this problem by finite difference method. Then, combining the vacuum rapid initial value, parameter homotopy, improved Newton iteration algorithm with relaxation factor to solve the problems of initial value conjecture and algorithm convergence difficulties. Secondly, a new multi-level fast solution strategy is proposed, which improves the solution velocity of the algorithm greatly and ensures the algorithm’s reliable convergence while solving for large-scale discrete interval precisely. Finally, the numerical simulation of the algorithm’s solution accuracy, efficiency, task adaptation is verified.A rapid trajectory optimization with regular perturbation method is researched. In order to avoid the indirect method’s drawback of improving solution precision by increasing the discreted interval numbers and decreasing the solution velocity, the regular perturbation method can realize the fast and high precision solution of atmospheric ascent section of launch vehicles’ three-dimensional trajectory optimization by using the efficiency of solving optimal control problems. The method devides the optimal trajectory problem into second optimal zero order model based on collocation method and higher order problem solution layered model by matching point error correction and compensation. In order to improve the efficiency of zero order problem solving, this paper constructs a vacuum based on analytical solution and numerical solution of atmospheric effects of hybrid algorithm, and puts forward a multi-objective fractional iteration fast solution strategy. For the complexity of the higher order differential equations of the model and the difficulty of accessing the state transition matrix for difficult problems, this paper designs the numerical differential equation model in advance reserve strategy and improve product distribution point algorithm.A hybrid optimization method based on the indirect method and Gauss pseudo-spectral method is proposed. The indirect and regular perturbation methods exist some problems including the accuracy, optimality and solving efficiency conflict. A kind of hybrid algorithms is proposed, applying the Gauss pseudo-spectral method to solve the HTPBVP. In the method, both the state variables and costate variables are discreted with the same interpolation functions. And the differential equations are replaced by the derivative of global interpolation polynomials in the Gauss quadrature formula. The HTPBVP is converted into algebraic equations with the method. The simulation results show that the hybrid optimization algorithm has better solving efficiency, accuracy and optimal performance index than the indirect method and the regular perturbation method.For the purpose of realizing the trajectory online planning and closed-loop guidance, in this paper a general autonomous trajectory online optimization method based on nonlinear programming algorithm is proposed. For the problems that solving the optimal algorithm of ascending trajectory through the maximum principle is complex, and with poor generality, this paper proposes a general nonlinear online optimization algorithm. The algorithm combined the online flight path prediction, MGPA two-step iterative nonlinear programming technology and instruction parameterization, path constraints control strategy, in order to satisfy the real-time requirements for online trajectory planning and guidance, and the concrete strategy of instruction parameterization is put forward. The process of the construction of the algorithm is versatile and can be applied to trajectory online planning and guidance tasks including reentry return flight. Parameterized instruction model is built, the optimal ascending trajectory in atmospheric online planning problems is embedded in the algorithm, simulation results show that the algorithm can satisfy the real-time and accuracy requirements, the feasibility and effectiveness of the algorithm are verified.Trajectory online planning and closed-loop guidance method of launch vehicles’ atmospheric ascent based on rapid trajectory optimization algorithms. With the realization of rapid trajectory optimization algorithms, the strategies of trajectory online planning, online reconfiguration and closed-loop guidance are studied to meet the requirements of real-time and flight safety. The strategies include the reasonable selection trajectory planning cycle, adaptive to solve the variable selection, online serial optimization, adaptive feedback update and strong path constraint. The numerical simulations under the normal condition and the condition of the engine failure are respectively made with the indirect method, regular perturbation method, hybrid optimization method and the nonlinear programming method. The results show that the proposed trajectory online planning and closed-loop guidance methods have the abilities of autonomous and high precision guide, reliable convergence, flight safety and trajectory online replanning when failure cases happen.This paper extends the research contents of trajectory design and guidance methods for launch vehicles, explores some methods of high solving precision and efficiency, versatility, new rapid trajectory optimization design and closed-loop methods. The relational research results in this paper have certain reference significance for the overall design and advanced guidance schemes of the advanced launch vehicles.