基于多模型预测的再入飞行器制导方法

在具有控制和轨迹约束的再入制导中,提出了一种新的基于多模型预测再入飞行器纵向制导方法。首先对再入飞行器的纵向运动方程沿着标准轨道线性化,基于分段仿射约束系统建模理论逼近原非线性系统,并采用线性矩阵不等式(LMIs)优化技术离线将保证闭环系统稳定的终端二次代价项求出;然后,在每个制导周期内,基于多模型预测在线求解构造好的有限时域优化目标,从而获得控制量的增量,并把第一个控制增量分量与标准轨道的控制量叠加后形成全量控制,用于实际再入轨道的制导,而在下一个制导周期基于新的状态参数重复上面的优化过程。数字仿真结果证实了所提方法的有效性。

Guidance Method Based on Multi-model Prediction for Re-entry Vehicles

A new method based on multi-model prediction is proposed for the longitudinal guidance of re-entry ve-hicles with trajectory and control constraints.Firstly,the linearized longitudinal dynamic equation of the nominal tra-jectory is modeled based on the constrained piecewise affine systems,which can model a wide range of practically relevant nonlinear systems.Using technique from theory of linear matrix inequalities(LMIs),the terminal cost matrix is given for closed-loop systems stability.Secondly,in each guidance period the control increment is obtained by solving on-line finite predictive objective based on multi-model predictive.Finally,the first control increment plus the nominal one is applied to guide the actual reentry trajectory,and the above procedure is repeated based on new measurements at the next period.Simulation results validate the effectiveness of the method.