提出了一种基于系统辨识获取飞行动力学模型并进行鲁棒飞行控制律综合设计的方法。首先利用混合遗传粒子群优化算法,从真实试飞数据中辨识建立无人机动力学模型,然后基于随机鲁棒原理的设计方法和辨识模型,设计了无人机鲁棒飞行控制律。飞行试验验证表明,在无气动数据情况下所设计出的高性能鲁棒飞行控制律,满足了无人机飞控研制的需求。 A method based on system identification for obtaining flight dynamics model and comprehensive design of robust flight control law is proposed. Firstly, hybrid GA-based particle swarm optimization algorithm was used to identify the UAV dynamics model from the real test data. Then based on the stochastic robust principle design method and identification model, a UAV robust flight control law was designed. Flight test verification shows that the robust control law of high performance designed in the absence of aerodynamic data meets the demand of UAV flight control development.