基于遗传算法和神经网络的倒立摆控制系统

以单级倒立摆为对象,介绍了一种融合遗传算法的神经网络控制方法。该方法采用以多层前馈神经网络作为遗传搜索表示方法的思想,以神经网络为基础,用遗传算法来学习神经网络的权系数,既保留了遗传算法的强全局随机搜索能力,又具有神经网络的鲁棒性和自学习能力。仿真结果证明:遗传算法和神经网络的结合,可兼有神经网络广泛映射能力和遗传算法快速全局收敛等性能。

Pendulum Control System Based on Genetic Algorithm and Neural Network

To control a single inverted pendulum, a kind of neural network control method is introduced,in which genetic algorithm and neural network are mixed. The notion of using the multi-layer forward neural network as the representation method of the genetic searching technique is intro15duced, and the weighs of neural network are trained by genetic algorithm. So the method remains the global stochastically searching ability of genetic algorithm and the robustness and self-learning ability of neural network. The simulation results show that extensive mapping ability of neural network and rapid global convergence of genetic algorithm can be obtained by combining genetic algorithm and neural network.