一种改进工业机械臂动力学参数辨识方法研究

A Modified Identification Method of Industrial Robot's Parameter

以六轴工业机械臂为研究对象,以牛顿-欧拉法为基础对机器人动力学特性进行建模,同时考虑了机器人运动时关节摩擦力与伺服电机转动惯量的影响。通过对六机器人的动力学方程进行线性化,并使用QR分解剔除对动力学方程贡献很小的分量,得到动力学模型的最小参数集。同时,使用有限项傅里叶级数作为激励轨迹并给出了激励轨迹优化目标函数,将机器人各关节的角度、速度与加速度限位作为优化的约束条件。在获取机器人运行激励轨迹的真机数据后,使用Butterworth滤波器对力矩数据进行进一步的平滑。最终,使用加权最小二乘法对动力学模型的最小参数集进行参数辨识,并使用测试轨迹进行模型的验证。实验结果表明,使用此方法所建立的动力学模型与真实力矩数据趋势性一致,误差在可接受的范围内,可以用于拖动示教等柔顺控制实验。

Based on the newton-euler method,the dynamic characteristics of the robot are modeled based on the industrial robot,and the influences of joint friction and rotational inertia of the servo motor are also considered.By linearizing the dynamic equation of the six robots and using QR decomposition to eliminate the components that contribute little to the dynamic equation,the minimum parameter set of the dynamic model is obtained.At the same time,the finite Fourier series is used as the excitation trajectories and various optimization objective functions of the excitation trajectories are given.After obtaining the real machine data of robot running excitation trajectory,Butterworth filter was used to further smooth the torque data.Finally,the weighted least square method is used to identify the minimum parameter set of the dynamic model,and the test trajectory is used to verify the model.The experimental results show that the dynamic model established by this method is consistent with the trend of real torque data,and the error is within an acceptable range,which has certain teaching and research value.