论文部分内容阅读
针对机器人这种不确定性的复杂非线性系统很难建立其精确的数学模型这一问题,提出一种基于自适应神经模糊推理(ANFIS)的方法对机器人系统进行建模。此方法将模糊推理和神经网络的学习能力有机地结合起来,并利用神经网络的学习机制自动地从输入输出数据中提取规则。建模过程中为了给ANFIS赋予一个合适的初始状态,选用减法聚类对输入数据进行处理。ANFIS网络的所有参数采用混合算法进行调节,即前提参数采用误差反向传播法,结论参数采用最小二乘法。最后在Matlab中对二自由度机器人进行仿真研究,仿真结果表明该方法模型结构简单,建模速度快,辨识精度高,同时也验证了该方法的有效性,为进一步实现机器人鲁棒自适应控制打下基础。
Aiming at the problem that it is very difficult to build a precise mathematical model for complex nonlinear systems with uncertainties such as robots, a method based on adaptive neuro-fuzzy inference (ANFIS) is proposed to model the robot system. This method combines the fuzzy reasoning and the learning ability of neural network organically, and uses the learning mechanism of neural network to automatically extract the rules from the input and output data. In order to give ANFIS a suitable initial state during the modeling process, subtractive clustering is used to process the input data. All parameters of the ANFIS network are adjusted by a hybrid algorithm, that is, the premise parameter adopts the error backpropagation method, and the conclusion parameters are least squares. Finally, the two-degree-of-freedom robot is simulated in Matlab. The simulation results show that the proposed method has the advantages of simple structure, fast modeling speed and high recognition accuracy. At the same time, the validity of this method is verified. In order to further realize Robust Robust Adaptive Control lay the foundation.