摘要
针对多自由度、高柔性的混凝土泵车臂架运动学逆解不唯一的问题,利用RBF神经网络来求解其逆解。在RBF神经网络的训练中合理利用正交最小二乘法,使得构建的RBF神经网络快速稳定。经MATLAB仿真分析表明,该方法计算量小且精度高,能将各臂节旋转误差控制在0.5°以内。
Aiming at the unique inverse kinematics solution for boom of concrete pump vehicles with the multi-degree and higher flexibility,the RBF neural network is used to solve the inverse solution,the Orthogonal Least Squares(QLS) method is used rationally,which makes the RBF neural network fast and stable.Finally,it is manifest that the method has a small amount of calculation and higher precision with using MATLAB for simulation analysis,which the angle of rotation can be reduced in the range of 0.5°.
引文
[1]黄毅,吴斌兴,王佳茜,等.混凝土泵车臂架振动响应的主动控制实验研究[J].振动与冲击,2012,31(2):91-94.
[2]刘杰,戴丽,赵丽娟,等.混凝土泵车臂架柔性多体动力学建模与仿真[J].机械工程学,2007,43(11):131-135.
[3]刘永超,黄玉美,王效岳,等.基于遗传算法的机器人运动学逆解[J].机器人,1998,20(6):421-426.
[4]MOODY J.DARKEN C.Fast learning in networks of locally tuned processing units[J].Neural Computation,1989,1(1):281-294.
[5]HAYKIN S.Networks:A Comprehensive Foundatio-n[M].2版.北京:清华大学出版社,2001.
[6]王小川,史峰.MATLAB神经网络43个案例分析[M].北京航空航天大学出版社,2013.
[7]CHEN S,COWAN C F N,GRANT P M.Orthogonal least squares learning algorithm for radial basis function networks[J].IEEETransactions on Networks,1991,2(2):302-309.
[8]罗小平,韦巍.一种基于生物免疫遗传学的冗余机械手轨迹规划新方法[J].模式识别与人工智能,2002,15(3):299-304.