混合小推力航天器轨道保持高性能滑模控制

针对采用太阳帆、太阳电混合小推力推进的航天器，研究了其在日心悬浮轨道的保持控制问题。为解决已有控制方法中未综合考虑内部未建模动态和外部未知扰动的问题，以及进一步提高系统控制性能，设计了一种高性能滑模控制策略。首先，考虑模型不确定性，建立了混合小推力航天器在日心悬浮轨道柱面坐标系的动力学方程；其次，基于改进型条件积分滑模面和径向基（RBF）神经网络设计了控制律，结合自适应方法在线估计不确定参数；接着，将求取的虚拟控制量在推进剂最优条件下转换成实际控制量，即太阳帆姿态角和太阳电推进力；最后，数值仿真验证了上述设计方法提高了系统鲁棒性，减小了轨道位置超调，并且混合推进相比于单一太阳帆推进，在更短收敛时间内控制精度提高了4个数量级，相比于单一太阳电推进，一年可以节省约89.6%的推进剂。

High-performance sliding mode control for orbit keeping of spacecraft using hybrid low-thrust propulsion

For a spacecraft using hybrid solar sail and solar electric propulsion, the station-keeping control of the heliocentric displaced orbit is investigated. To solve the problem that internal unmodeled dynamics and external unknown disturbances are not considered comprehensively in the existing methods, and to further improve the performance of the system, a high-performance sliding mode control strategy is designed. Firstly, considering the uncertainty of the model, the dynamic equation of the hybrid low-thrust spacecraft keeping on heliocentric displaced orbit is established in the cylindrical coordinate system. Secondly, the control law is designed based on the improved conditional integral sliding surface and Radial Basis Function (RBF) neural network, and the uncertain parameters are estimated online by combining the adaptive method. Then, under the optimum condition of propellant, the virtual control variables are converted into actual control variables, namely attitude angles of solar sail and solar electric propulsion. Finally, numerical simulation verifies that the above design enhances the robustness of the system, reduces the overshoot of orbit position, and hybrid propulsion improves the control accuracy by 4 orders of magnitude in shorter convergence time compared to single solar sail propulsion, while it can save about 89.6% propellants a year compared to single solar electric propulsion.