基于人工神经网络的机械臂运动学分析

Mechanical Arm Kinematics Analysis Based on Artificial Neural Network

随着现代科技的日新月异,智能机械臂在实际生产制造中的应用越来越广泛。为了更加清晰地观测机械臂末端运动轨迹,实现直观地看到路径规划结果的特殊功能,就需要提升逆运动学求解的速度和精度。常规的方法是使用BP神经网络进行研究,但是这种方法存在收敛速度慢、局部极小值频出等问题。就针对这些问题改进了算法,实验结果表明该改进算法在实际应用中是可行的,并且该算法不仅能够提升收敛速度,还能提高逆运动学求解的准确率,可以较好应用于机械手逆运动学求解的问题。

With the rapid development of modern science and technology,intelligent robotic arms have become more and more widely used in practical manufacturing.In order to observe the movement trajectory of robot arm more clearly and achieve the special function of visualizing the result of path planning,the speed and accuracy of inverse kinematics solution need to be improved.The conventional method is to use BP neural network to study,but this method also has some problems such as slow convergence rate and local minimum frequency.In this paper,the algorithm is improved for these problems,the experimental results show that the improved algorithm is feasible in practical application,and the algorithm can not only improve the convergence rate,but also improve the accuracy of inverse kinematics,which can be better applied Inverse kinematics problem of manipulator.