论文部分内容阅读
驾驶员操纵行为受到自身、外界环境和被控对象等多方面的影响,因此驾驶员模型具有非线性的特征。神经网络模型克服了拟线性模型不能反映驾驶员非线性操纵的问题。为了获得建立模型的数据,利用地面模拟器使飞行员对一系列指令进行精确跟踪。获得的指令、飞机状态和驾驶员输入信息等参数即可作为神经网络模型的训练样本。神经网络驾驶员模型的训练和测试结果表明,该建模方法是合理、准确的,可以应用于人机闭环系统中驾驶员操纵的研究。
The driver’s behavior is influenced by many factors such as itself, external environment and controlled object, so the driver’s model has nonlinear characteristics. The neural network model overcomes the problem that the quasi-linear model can not reflect the driver’s nonlinear manipulation. To get model-building data, a ground simulator is used to enable the pilot to accurately track a series of instructions. The acquired instructions, aircraft status and driver input information can be used as training samples for the neural network model. The training and testing results of the neural network driver model show that this modeling method is reasonable and accurate and can be applied to the research of driver control in man-machine closed-loop system.