模糊神经网络的二级倒立摆稳定控制分析

Equilibrium control analysis of double-link inverted pendulum based on fuzzy-neural-network

在MATLAB环境下,针对二级倒立摆系统稳定控制问题,引入新的智能控制策略,该种方法采用BP算法与最小二乘(LSE)算法结合的混合算法对Takagi-Sugeno模糊模型中的前件及后件参数进行优化修正,在已获得的客观输入、输出样本集的基础上,提出一种基于自适应神经网络的模糊推理系统ANFIS来对倒立摆系统进行"倒立"控制。实时控制结果表明,所提出的控制方法是可行而且有效的。

Under the environment of MATLAB, aiming at equilibrium control problem of double-link inverted pendulum system, a new intelligent control strategy is introduced. A mixed arithmetic of BP and LSE arithmetic are used, in order to optimize and amend the front-part and later-part parameter of Takagi-Sugeno fuzzy model. Based on the objective input-output sample set acquired, a sort of self-adaptive neuro-fuzzy inference system is put forward for inverted control of inverted pendulum. The result of real- time equilibrium control shows this control method is available and effective.