摘要
针对可重复使用火箭回收垂直着陆阶段环境条件复杂、精度要求严苛,以及模型存在不确定性的问题,设计了一种自适应模型预测控制器,为垂直着陆段高精度抗扰控制难题提供解决方案。该方法在模型预测控制滚动优化的基础上,用预测模型响应与实际系统响应的误差驱动底层PD控制器参数更新,使得实际状态收敛于预测状态,补偿模型不准确带来的偏差。仿真结果表明:针对复杂的火箭垂直着陆控制任务,所提出的控制方法在模型存在30%的结构参数偏差下动态响应质量较好,且相较于传统模型预测控制方法具有更高的落点控制精度和鲁棒性。
To address the challenges of complex environmental conditions,stringent accuracy requirements,and uncertainties in the vertical landing phase of rocket recovery,an adaptive model predictive controller is designed to provide a solution for the high-precision disturbance rejection control problem during vertical landing.Based on the receding horizon control of model predictive control,this method updates the parameters of the underlying PD controller using the error between the predicted model response and the actual system response,ensuring that the actual state converges to the predicted state and deviations caused by model inaccuracies is fixed.Simulation results demonstrate that,with a 30%deviation in structural parameters,the proposed approach maintains favorable dynamic response characteristics and achieves higher landing-point accuracy and robustness compared with conventional MPC methods in complex rocket vertical landing tasks.
作者
于千惠
葛亚杰
张冉
李惠峰
YU Qianhui;GE Yajie;ZHANG Ran;LI Huifeng(School of Astronautics,Beihang University,Beijing 102206,China;Key Laboratory of Spacecraft Design Optimization and Dynamic Simulation Technologies,Ministry of Education,Beijing 100191,China;Science and Technology on Space Physics Laboratory,Beijing 100076,China)
出处
《飞行力学》
北大核心
2025年第6期63-69,共7页
Flight Dynamics
