论文部分内容阅读
将函数联接型神经网络（ＦＬＡＮＮ）引入传感器动态特性的研究，利用神经元网络良好的逼近能力，建立腕力传感器的动态数学模型．该方法所建模型阶次低、精度高，对数据个数和采样频率无特殊要求．根据“逆模型”的思想，提出了基于函数联接型神经网络的传感器动态补偿方法．此方法设计出的动态补偿器简单、实时性好；不依赖于传感器的模型，鲁棒性强．
The function-coupled neural network (FLANN) is introduced into the study of the dynamic characteristics of the sensor. The dynamic approximation ability of the neural network is used to establish the dynamic mathematical model of the wrist force sensor. The method has low modeling order and high precision, and has no special requirements on the number of data and the sampling frequency. According to the idea of “inverse model”, a dynamic sensor compensation method based on function-linked neural network is proposed. The dynamic compensator designed by this method is simple and has good real-time performance. It does not depend on the sensor model and has strong robustness.