论文部分内容阅读
研究了机械手控制的逆运动学问题，提出了一种克服Ｈｏｐｆｉｅｌｄ网络的局部极值问题的网络参数扰动算法，通过数字模拟分析了该算法的性能，并将该算法成功地应用于机械手控制的逆运动学问题。计算机仿真表明，这种神经网络控制方法不仅具有较快的速度，而且大大提高了对机械手的控制精度。
The inverse kinematics problem of manipulator control is studied. A network parameter perturbation algorithm to overcome the local extremum problem of Hopfield network is proposed. The performance of the algorithm is analyzed by digital simulation. The algorithm is successfully applied to manipulator inverse Kinematics problems. Computer simulation shows that this neural network control method not only has a faster speed, but also greatly improves the control accuracy of the manipulator.