摘要
首先通过Lagrange法构建机器人的动力学模型,经过线性变换和动力学参数重组,获取最小惯性参数集。其次,选择傅里叶级数作为激励轨迹,结合MATLAB优化工具箱对傅里叶级数系数进行优化,从而获得最佳的激励轨迹。最后将最优的激励轨迹导入到ADAMS中进行仿真,得出相关数据,并根据最小二乘法计算出待辨识动力学参数估计值,并与理论值进行对比,计算绝对误差。结果表明:仿真实验能够获得较为理想的效果,验证参数辨识方法的正确性。
First of all, the dynamic model of the robot is built according to the Lagrange method, and the min-imum set of unbiased parameters is obtained after linear transformation and dynamic recombina-tion of parameters. Secondly, the Fourier series is chosen as the exciting trajectory, and the Fourier series coefficient is optimized in conjunction with the MATLAB optimization tools to achieve the best exciting trajectory. Finally, the best gripping trajectory is imported into ADAMS for modeling, the relevant data is obtained, and the calculated value of the known kinetic parameter is calculated by the method of least squared, and the total error is calculated by comparison with the theoretical value. The results show that by correcting the parameter identification method and verifying the parameter identification method, a more ideal effect can be achieved.
出处
《建模与仿真》
2023年第2期1428-1440,共13页
Modeling and Simulation
