基于遗传神经网络求解机械手逆运动学问题  

Study on Control Problem of the Inverse Kinematics of Robots Based on Neural Networks and Genetic Algorithm

作　　者：王涛[1] 季英瑜[1] 张庆功[1] 

机构地区：[1]浙江工业大学浙西分校,浙江衢州324000

出　　处：《轻工机械》2007年第5期39-41,共3页Light Industry Machinery

摘　　要：首先采用神经网络逆辨识模型建立机械手的逆运动学模型,结果显示该法不可行,然后采用神经网络模型和遗传算法相结合的方法来解决其逆运动学控制问题,结果表明此方法即便是在系统不存在其逆映射时,也能获得很好的效果,从而验证了该方法的有效性。At first, the inverse neural network is used to construct the inverse kinematics of robots, but the results show that this method is unsuccessful. In the end, the new method based on neural network and genetic algorithm is applied to solve the control problem of the inverse kinematics of robots, and the results show that even though the system doesn＇t exist the inverse mapping, the accurate result can be gotten. As a result, the proposed method is proved to be practical and valid.

关 键 词：神经网络 遗传算法 机械手 逆运动学 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TP241[自动化与计算机技术—控制科学与工程]