基于改进序列凸优化的固体火箭入轨制导方法  

Orbital guidance method for solid rocket based on improved sequential convex optimization

作　　者：王洪波[1] 张若男 梁卓[1] 吕瑞 WANG Hongbo;ZHANG Ruonan;LIANG Zhuo;LYV Rui(China Academy of Launch Vehicle Technology,Beijing 100076,China)

机构地区：[1]中国运载火箭技术研究院,北京100076

出　　处：《中国惯性技术学报》2021年第1期119-125,共7页Journal of Chinese Inertial Technology

基　　金：装备发展领域基金(41412050)。

摘　　要：针对推力大小与关机时间不可调节的耗尽关机固体火箭入轨制导问题,提出一种改进序列凸优化算法。首先利用Legendre-Gauss-Radau伪谱法对模型离散,将滑行时间增广为控制变量,解决点火时间不确定问题,保证求解最优性。其次对非凸推力矢量约束进行松弛凸化,并证明松弛手段的无损性;然后,对非凸运动方程与终端状态约束进行线性化,构建凸优化模型。最后设计初值生成器,为快速迭代构建良好的初始轨迹。仿真结果表明,该制导算法能够满足高精度入轨要求,且问题求解时间仅为伪谱法的6.72%,验证了改进序列凸优化算法有效性。为耗尽关机固体火箭入轨在线高精度快速制导提供重要参考。An improved sequence convex optimization algorithm is proposed for orbital guidance problem of a exhausted rocket with un-adjustable thrust and shutdown time.Firstly,the Legendre-Gauss-Radau pseudo spectral method is used to discrete the model to increase the sliding time to a control variable to solve the problem of uncertain ignition time and ensure the solution optimality.Secondly,the non-convex thrust vector constraint is relaxed convex and the lossless property of the relaxation method is proved.The non-convex motion equations and terminal state constraints are linearized to build a convex optimization model.Finally,an initial value generator is designed to quickly construct a good initial trajectory.Simulation results show that the guidance algorithm can meet the requirements of high precision orbit injection,and the problem-solving time is only 6.72%of the pseudo-spectral method,which verifies the effectiveness of the improved sequential convex optimization algorithm,and provides an important reference for the online high-precision and rapid guidance of the exhausted shut-off solid rocket orbital.

关 键 词：序列凸优化 固体火箭 耗尽关机 入轨制导 

分 类 号：V279[航空宇航科学与技术—飞行器设计]