基于GA-SVM优化算法的扒渣机器人逆运动学求解研究

Inverse kinematics of slag raking robot based on GA-SVM optimization algorithm solving research

扒渣机械臂逆运动学分析是其末端破碎锤与铲斗实现运动轨迹精准控制的难点,由于封闭解法通用性差,以及运用单一神经网络求取时存在精度不足等问题,提出一种基于遗传算法(GA)优化支持向量机(SVM)的机械臂逆运动学求解方法。首先由D-H参数法构建正运动学模型,并获取数据样本;再采用GA算法对SVM的惩罚因子c和核函数参数g以及损失参数ε进行全局寻优,将构建的GA-SVM模型应用到扒渣机械臂关节角逆运动学求解中,结果显示末端位姿的均方误差为1.54026×10^(-7),表明该算法可准确求解出期望关节角度值;最后将算法模型求取的逆解数据进行五次多项式轨迹规划,验证了GA-SVM算法模型的可靠性。

The inverse kinematics analysis of the slag raking robotic arm is a difficult point for its end crushing hammer and bucket to achieve accurate control of the movement trajectory.Due to the poor generality of the closed solution method and the lack of accuracy when using a single neural network to solve the problems,a method of inverse kinematics solving for the robotic arm based on the Genetic Algorithm(GA)optimized Support Vector Machine(SVM)is proposed.Firstly,the positive kinematics model is constructed by D-H parameter method and the data samples are obtained;then the GA algorithm is used to optimize the penalty factor“c”and kernel function parameter“g”of the SVM as well as the loss parameter“ε”globally,and then the constructed GA-SVM model is applied to inverse kinematics of the joint angle of slag raking robotic arm,and the results show that the mean square error of the end position is 1.54026×10^(-7),which indicates that the algorithm can accurately solve the desired joint angle.Finally,the inverse solution data solved by the algorithmic model is subjected to five times polynomial trajectory planning to verify the reliability of the GA-SVM algorithmic model.