【作者】 雍恩米；

【导师】 唐国金；

【作者基本信息】 国防科学技术大学 ， 航空宇航科学与技术， 2008， 博士

【摘要】 高超声速滑翔式再入飞行器具有较大的机动能力,是实现远程快速精确打击或力量投送的新型再入飞行器。本文以解决高超声速滑翔式再入飞行器轨迹优化与制导关键技术为目标,系统研究了再入弹道特性和最优飞行攻角方案、再入轨迹优化、轨迹在线生成和制导以及弹道突防能力分析问题,主要研究成果如下:基于平面再入运动模型,研究了高超声速滑翔式再入飞行器的基本再入弹道形式与最优攻角方案。1)分析了再入点参数和最大升阻比对弹道的影响;2)应用物理规划方法求解了多目标最优再入轨迹,分析了沿最优轨迹飞行的物理原因;3)进一步提出了参数化最优攻角方案。提出了基于Gauss伪谱法的高超声速滑翔式再入飞行器快速轨迹优化方法。针对传统轨迹优化方法在初值选取、计算效率上的不足,提出了基于Gauss伪谱法的含初值生成器的分段串行优化策略。应用该优化策略计算了到达目标点的最优再入轨迹、到达目标区域的最优再入轨迹和再入飞行的可达区。通过与数值积分解比较,验证了方法的有效性。在一般高性能微机上,计算约一万公里航程的最优再入轨迹与再入飞行近似可达区的时间均为3分钟左右,体现了方法的快速性。改进了远程多约束三自由度再入轨迹在线生成方法。1)建立了给定飞行攻角方案和约束条件下的再入走廊数学模型;2)提出并利用改进的拟平衡滑翔条件,将弹道约束转换为控制变量约束,并将控制变量参数化,再利用速度与航程的近似关系,求解满足终端约束的控制变量参数和纵向参考轨迹;3)利用线性二次调节器跟踪纵向参考轨迹,引入航向角误差走廊控制横向轨迹,完成三自由度轨迹在线生成;4)仿真验证了该方法在一般高性能微机上耗时2-3s即可生成航程约一万公里的满足多约束的再入轨迹。提出了设定航路点的预测-校正与跟踪控制的混合制导策略。将航迹分为预测-校正段和接近目标的重新瞄准段,前一段在设定的航路点间制导,后一段在线重生成参考轨迹并跟踪制导。1)提出了航路点确定方法;2)预测-校正段将控制变量参数化,基于大脑情感回路智能控制器建立了航路点间的预测-校正制导算法;3)重新瞄准段在线重生成轨迹,并利用线性二次调节器实现轨迹跟踪;4)仿真验证了标准条件、考虑再入点参数、大气密度、飞行器质量、升力系数和阻力系数偏差条件和目标机动条件下混合制导方法是有效的,且与一般预测-校正制导方法比较,混合制导策略在制导精度、在线预测时间等方面具有优势。研究了雷达对常规再入弹道和滑翔式再入弹道的跟踪效果。基于Unscented卡尔曼滤波和考虑气动模型的目标跟踪模型,加入雷达测量噪声,计算目标跟踪误差,比较不同形式弹道的跟踪效果。仿真结果表明,滑翔式再入弹道由于其机动特性,相对弹道式再入在突破雷达跟踪上具有优势。论文拓展了飞行器轨迹优化数值方法的研究范畴,探索了再入制导方法的新思路,引入了智能控制领域的新方法,具有一定理论意义;同时,研究充分结合高超声速滑翔式再入飞行器的应用背景,基于理论研究成果分析其再入弹道特点与性能,对发展未来新型高超声速滑翔式再入飞行器具有重要借鉴意义。更多还原

【Abstract】 The hypersonic glide-reentry vehicle is a new maneuverable vehicle for the long-range precise strike mission or military delivering. For the purpose of developing reentry trajectory optimization and guidance technology of the hypersonic glide-reentry vehicle, this dissertation studies the properties of the reentry trajectory, the schematic of reentry angle-of-attack, trajectory optimization, on-board trajectory generation, reentry guidance and the penetration capability of glide-reentry. The main results achieved in this dissertation are summarized as follows:Based on the planar reentry dynamic model, the rudimental properties of the glide-reentry trajectory and the angle-of-attack scheme are studied. 1) The influence of the initial trajectory parameters and the maximum lift-to-drag ratio on the trajectory shape is analyzed; 2) The physical programming method is employed to solve multi-objective reentry trajectory optimization problem, and the physic characteristic of the optimal trajectory shape is analyzed; 3) Then the optimal parameterized angle-of-attack scheme of reentry is proposed.A rapid trajectory optimization approach using Gauss Pseudospectral Method (GPM) for hypersonic glide-reentry vehicle is developed. Aimming at deficiencies of traditional trajectory optimization method in initial value determination and computation efficiency, a pipelining and segmenting trajectory optimization approach base on GPM, containing an initial guess generator, is proposed. Then this appoach is applied to compute the optimal trajectories of reaching a point target, an area target and the attainable region of reentry flight. The feasibility of this approach is validated by comparing the GPM results with that of the numerical integration. An optimal trajectory with a range of ten thousand kilometer and the attainable region of reentry flight both can be generated in about 3 minuets on a desktop computer. It indicates the rapidity of the presented approach.An improved on-board three-degree-of-freedom (3DOF) constrained entry trajectory generation method is proposed. 1) With the defined angle-of-attack profile and path constraints, the mathematical model of reentry corridors are biult; 2) An improved qusi-equilibrium glide condition is presented and utilized to convert all path constraints to the constraints of control variables, and the control variables are parameterized. Employing the approximate relationship between the velocity and the range, the parameters of control variable and the longitudinal reference trajectory satisfying terminate constraints can be solved; 3) Then the 3DOF trajectory generation method is completed by the heading error corridor control strategy in lateral motion and longitudinal trajectory tracking control based on the linear quadratic regulator (LQR); 4) The numerical simulation indicates that the algorithm is able to generate a reentry trajectory with a range of ten thousand kilometer satisfying all path constraints in about 2-3 seconds on a desktop computer.A mixed reentry guidance of predictor-corrector with defined way-points and trajectory tracking is proposed. In the solution process, the reentry trajectory is divided into the predictor-corrector part and the retargeting and tracking part which is near to the target. In the first part, the predictor-corrector method is applied to solve the guidance problem between the defined way-points. In the second part, reference trajectory is regenerated on-board and tracked. 1) The method of way-points determination is presented. 2) In the predictor-corrector part,the control variables are parameterized and the predictor-corrector algorithm is developed using the brain emotional learning based intelligence controller. 3) In the retargeting part, the trajectory is regenerated on-board and the LQR theory is employed for trajectory tracking. 4) The effectivity of the mixed guidance strategy is validated by simulations in conditions of the nominal case, the dispersed case, which consider deviations of reentry point parameters, air density, vehicle mass, lift coefficients and drag coefficients, and the target moving case. Comparing with the general predictor-corrector guidance method, the mixed guidance strategy has advantages in guidance precision and the time cost of trajectory predictionThe radar tracking performance for conventional reentry trajectory and the glide-reentry trajectory is studied respectively. Based on the Unscented Kalman Filter and target dynamics model, the tracking error considering the measure yawp of radar is calculated and the tracking performance is compared. The simulation results indicate that the glide-reentry trajectory excels the conventional trajectory in breaking the radar tracking.This dissertation has some theoretical significance in extending the research domain of current numerical trajectory optimization method, exploring a new idea of reentry guidance, and importing a new controller of intelligence control. The obtained properties and performance of glide-reentry trajectory based on theoretical researches are a good reference to the development of the new type of hypersonic glide-reentry vehicle in the future.更多还原