机械手逆运动学神经网络建模与仿真

Neural Network Modeling and Simulation for Inverse Kinematics of Manipulator

采用多层前向神经网络来建立机械手逆运动学模型。在分析了用简单遗传算法学习神经网络存在求解速度慢、精度低及有量化误差等缺点的基础上 ,提出采用改进遗传算法来学习神经网络 ,用于机械手逆运动学求解。此方法采用实数值编码 ,并采取动态变异操作。仿真结果表明 ,该方法有效地弥补了简单遗传算法的不足 ,能快速达到全局收敛 ,从而大大提高了机械手逆运动学求解的精度。

A kind of multilayer forward neural network is applied to build the inverse kinematics model of manipulator.The simple genetic algorithm has some shortcomings,such as low speed of solving problem,low precision and existing of quantization error.In this paper,a kind of improved genetic algorithm is proposed to learn neural network,which is used to solve the inverse kinematics of manipulator.Floating encoding and dynamic mutation are applied in this method.The simulation results show that the method can overcome deficiencies of simple genetic algorithm effectively and achieve the global convergence rapidly,therefore it greatly improves the precision of solving inverse kinematics of manipulator.