伴随方法在亚轨道飞行器预测制导中的应用

Rapid optimal predictive guidance via enhanced adjoint method for suborbital launch vehicles

摘要:

提出了一种求解最优两点边值问题的改进型伴随方法,并将其应用于一类以固体火箭发动机为动力的亚轨道飞行器轨迹优化问题中,间接得到一组平滑控制量;利用标称轨迹线性化概念,对该飞行器在标称轨迹附近做一阶泰勒展开,在给定可用姿态指令和摄动、干扰的上界条件下,利用滚动时域控制技术,得到了在系统摄动和干扰存在情况下拥有高精度、快速鲁棒性和良好动态响应品质的跟踪结果,同时具有一定的预测校正能力.以某高速飞行器为例,对该系统进行最短时间优化,并在系统建模不确定存在的情况下,通过滚动时域控制技术对最优标称轨迹进行跟踪,仿真结果证明该组合方法有效、可靠.

Abstract:

An enhanced adjoint method for solving two point boundary value problems is proposed and applied to trajectory optimization problems for a rocket based high speed suborbital launch vehicle,and smooth control serials are obtained indirectly.The first order Taylor Expansion is used to the nominal trajectory which is calculated by the enhanced adjoint method.High robust,high accurate tracking result is obtained under disturbance boundary constrains by receding horizon control technology which is featuring prediction ability.The simulation results demonstrate the capacity and reliability of this proposed method.