基于微分方程数值分析的机械臂运动参数挖掘模型

在目前的运动参数挖掘模型中,通常采用差异化运动原理对运动过程进行分析,缺乏实际的公式支持,在进行参数挖掘时影响性能。因此,提出基于微分方程数值分析的机械臂运动参数挖掘模型研究。首先利用微分方程数值分析的方式来对机械臂的运动过程进行模拟,列举出其中的动态特性参数。依据动态特征参数,利用神经元网络对其中的运动参数属性进行特征提取。获得运动特征后,使用贝叶斯原理进行深入的运动参数挖掘。为了验证设计的运动参数模型性能,设计仿真实验,设定模拟机械臂,并使用设计模型与文献3]、文献5]、文献13]中的参数挖掘模型进行运动参数挖掘,并对比性能的优劣性。实验结果证明,设计的运动参数挖掘模型具有较高的运算速度、实时性以及准确率,证明设计的运动参数挖掘模型具有较高的性能,满足设计初衷。

Mining model of manipulator motion parameters based on numerical analysis of differential equation

In the current motion parameter mining model,the principle of differential motion is usually used to analyze the motion process,which is lack of practical formula support,which affects the performance of parameter mining.Therefore,based on the numerical analysis of differential equation,the mining model of manipulator motion parameters is proposed.Firstly,the motion process of the manipulator is simulated by numerical analysis of differential equation,and the dynamic characteristic parameters are listed.According to the dynamic characteristic parameters,the neural network is used to extract the feature of the motion parameters.After obtaining the motion features,the Bayesian theory is used to mine the motion parameters.In order to verify the performance of the designed motion parameter model,the simulation experiment is designed,and the simulation manipulator is set up.The design model is used to mine the motion parameters with the parameter mining models in references3],5]and13],and the advantages and disadvantages of the performance are compared.The experimental results show that the design of the motion parameters mining model has a high speed,real-time and accuracy,which proves that the design of the motion parameters mining model has high performance,and meets the original intention of the design.