基于伪谱重构的采样反馈近最优制导

Sampled-data feedback control for near optimal guidance via pseudospectral trajectory reconfiguration

利用采样反馈控制和轨迹快速重构技术,设计了固定采样反馈和自适应采样反馈两种有限推力远程变轨近最优闭环制导策略。建立了空间变轨和交会最优制导数学模型。结合变轨运动方程特征,给出了伪谱优化参数缩减方法和实时性提升策略;基于采样反馈和最优控制理论,利用采样数据进行连续轨迹重构,并将开环最优解进行闭环反馈以更新制导指令。仿真结果表明,两种策略在保证任务指标近最优性的同时,可以有效抑制地球扁率J2摄动和计算误差的影响;自适应采样策略自主性好、制导精度误差收敛快,但计算量和燃耗偏大,二者使用时需要根据具体任务要求合理选择。

Two closed-loop near optimal guidance schemes, which are based on sampled-data optimal feedbackcontrol and pseudospectral trajectory reconfiguration are studied and proposed for long-range rendezvouswith finite thrust. They are defined respectively as the fixed sampling period（F SP ） feedback and adaptive samplingperiod（A S P ） feedback schemes according to different sampling period. F irstly, the dynamics model andfuel-optimal guidance problem are described. Combining the characteristics of the dynamics model with thepseudospectral method, a computational efficiency refinement method with the reduced state parameters ？ is proposedto improve the efficiency of trajectory optimization. Starting from the sampled-data optimal feedback controltheory, the control command for closed-loop guidance can be updated and fed back continuously throughsuccessive trajectory reconfiguration using sampled data. The Simulation results indicate that the algorithms cannot only supply excellent near optimal solution for practical mission, but can manage numerical errors, disturbances,and the J 2 perturbation. Furthermore, the adaptive sampling period guidance scheme is not only moreautonomous ,but it has a better convergence for guidance accuracy with higher computation and fuel expense.Consequently, a favorable decision should be made for two schemes in application.