Research on Stochastic Robust Attitude Control Technology Based on Latin Hypercube Sampling

Liantao Xu,Yongji Wang,Lei Liu,Zhongtao Cheng,Ye Yang
DOI: https://doi.org/10.1109/iciea.2019.8833748
2019-01-01
Abstract:The traditional stochastic robust method, which uses Monte Carlo to estimate the probability, has the disadvantages of sampling point accumulation and long-time consumption. The Latin hypercube sampling (LHS) is used to sample the uncertainty domain, and the sampling time can be reduced. Then, the stochastic robust objective function is established, and the parameters of the control law are optimized by genetic algorithm, so that the control precision can meet the design target in the uncertain domain. The simulation results of a hypersonic vehicle show that the scheme can meet the system control requirements and has strong robustness.
What problem does this paper attempt to address?