基于PSO-RBF神经网络的串联机械臂逆运动学分析

针对目前基于神经网络对串联机械臂求逆解方法中出现的精度不足和实时性较差的问题,使用粒子群优化(particle swarm optimization,PSO)算法对径向基函数(radial basis function,RBF)神经网络进行结构优化,提出一种基于PSO-RBF神经网络的机械臂逆运动学算法。首先由正运动学模型获取神经网络训练和测试参数样本,经过欧拉角变换在神经网络输入端建立机械臂关节位姿映射关系,然后通过PSO算法对径向基核函数进行参数寻优并对测试样本求解分析,最后获取经逆运动学求解后机械臂的运动轨迹,验证了该算法的可靠性。仿真结果显示,由PSO-RBF神经网络逆运动学算法能够快速得出满足精度要求的关节角度,为进一步机械臂工业控制提供了理论支持。

Inverse kinematics analysis of series manipulators based on PSO-RBF neural network

In order to solve the problem of inaccuracy and poor real-time performance in the inverse solution method of series manipulator based on neural network, the particle swarm optimization algorithm is used to optimize the structure of radial basis function neural network. A manipulator inverse kinematics algorithm based on PSO-RBF neural network is proposed. Firstly, the neural network training and test parameter samples are obtained by the forward kinematics model. The Euler angle transform is used to establish the manipulator joint pose mapping relationship at the input of the neural network. Then the PSO algorithm is used to optimize the parameters of the radial basis kernel function and the test sample is solved and analyzed. Finally, the motion trajectory of the manipulator solved by inverse kinematics is obtained and the reliability of the algorithm is verified. The simulation results show that the inverse kinematics algorithm based on PSO-RBF neural network can quickly obtain the joint angle that meets the accuracy requirements, which provides theoretical support for further mechanical control of the manipulator.