基于人工神经网络BP算法的倒立摆控制研究

基于人工种经网络BP算法的倒立摆小车实验仿真训练模型，其倒立摆BP网络为4输入3层结构．输入层分别为小车的位移和速度、摆杆偏离铅垂线的角度和角速度．隐含层单元数16个．输出层设置为1个输出单元．输入层采用Tansig函数，隐含层采用Logsig函数，输出层采用Purelin函数．用Matlab6．5数值计算软件对模型进行学习训练，并与模糊控制逻辑算法对比，表明倒立摆控制BP算法精度高、收敛快，在非线性控制、鲁捧控制等领域具有良好的应用前景。

Research for Inverted Pendulum Control Based on BP Algorithm of ANN

The training model of test simulation for car of inverted pendulum based on BP algorithm of artificial neural networks (ANN) is a BP network that has 4-input and 3-layer structure. Input layer is respectively the displacement and speed of car, the angle and angle speed between pendulum bar and vertical line. Hidden layer has 16 units, output layer has an output unit. Tansig function is used in input layer, Logsig function is used in hidden layer and Purelin function is used in output layer. The model is learned and trained with Matlab 6.5 calculating software, fuzzy logic algorithm is compared, the experiment results proves BP algorithm for inverted pendulum control has hither precision, better astringency and lower calculation. This algorithm has wide application on nonlinear control and robust control field in particular.