洗出算法中滤波器的自然截止频率参数作为主要因素影响着洗出效果,进而影响飞行模拟器的运动逼真度,为得到滤波器自然截止频率最优参数,提出一种改进的多目标灰狼算法,旨在寻找最合适的滤波器参数。利用Logistic-tent映射进行初始化来提升种群的多样性,将差分进化算法的思想引入到多目标灰狼算法中进行迭代寻优,并采用非线性控制参数及引入惯性权重策略更新种群位置,有效地兼顾了算法的全局搜索和局部开发能力。通过以洗出算法的三个评价指标来建立目标函数,并利用模糊隶属度函数得到最优解。仿真及实验结果表明经过改进的多目标灰狼算法相比于经典洗出算法缩小了1.23s相位延迟并修正了感觉峰值,同时节省了18.5%运动平台的工作空间,洗出效果更加理想。

The natural cutoff frequency parameter of the filter in the wash-out algorithm is the main factor affecting the wash-out effect, and then affect the motion fidelity of the flight simulator. In order to obtain the optimal parameters of the natural cut-off frequency of the filter, an improved multi-objective grey wolf algorithm is put forward to find the most suitable filter parameters. The Logistic-tent mapping is used for initialization to improve the diversity of the population. The belief of the differential evolution algorithm is combined to the multi-objective grey wolf algorithm for iterative optimization. The nonlinear control parameters and the introduction of the inertia weight strategy are used to update the population position, which effectively balances the algorithm"s global search and sectional development abilities. The objective function is built by using three evaluation indicators of the wash-out algorithm, and use the fuzzy membership function to get the best solution. The simulation and experimental results demonstrate that the improved multi-objective grey wolf algorithm reduces the 1.23s phase delay and corrects the sensory peak compared with the classic wash-out algorithm, while saving 18.5% of the working space of the motion platform, and the wash-out effect is more ideal.

TP391???????

国家自然科学基金项目（面上项目，重点项目，重大项目）