摘要
针对运动轨迹具有重复性特点的多自由度机械臂,为实现机械臂轨迹的精确跟踪控制目的,以一种三自由度空间机械臂为研究对象,采用混合多项式插值法规划关节空间轨迹,提出了一种增益自适应整定的迭代学习控制方法,对机械臂关节轨迹进行了跟踪控制。首先,根据机械臂逆运动学分析,求解了运动过程中关键点对应的各关节角度,采用3-5-3多项式混合插值方法进行了轨迹规划;然后,采用拉格朗日方法建立了三自由度机械臂的动力学模型,在定常比例-微分(PD)型闭环迭代学习控制的基础上,提出了一种增益自适应整定的迭代学习控制方法,并使用Lyapunov稳定性理论对设计的迭代学习控制器进行了收敛性分析;最后,为验证增益自适应整定的迭代学习控制方法的有效性与优越性,在MATALB/Simulink中搭建了机械臂控制系统进行了仿真分析,并将其结果与采用定常PD迭代学习控制方法的结果进行了对比。研究结果表明:随着迭代次数的增加,机械臂关节轨迹的跟踪效果有显著提升;第5次迭代时,对机械臂关节角度轨迹和末端路径的跟踪达到了预期效果;第10次迭代后,机械臂关节角速度轨迹也能准确地跟踪期望轨迹;与定常PD型闭环迭代学习控制相比,增益自适应整定的迭代学习控制在跟踪机械臂关节轨迹时达到期望轨迹的迭代次数更少、学习速度更快、轨迹跟踪效果更好。该研究为有重复运动轨迹的机械臂提供了一种有效的控制方法。
For a multi-DOF robotic manipulator with repetitive motion trajectory characteristics,in order to achieve precise trajectory tracking control,a 3-DOF spatial robotic manipulator was taken as the research object,the joint space trajectory was planned by the mixed polynomial interpolation method,and the gain adaptive tuning iterative learning control method was proposed to track the trajectory of the robotic manipulator.Firstly,according to the inverse kinematics analysis of the robotic manipulator,the angle of each joint corresponding to the key points in the motion process was solved.A 3-5-3 polynomial mixed interpolation method was used for trajectory planning.Then,dynamic model of the 3-DOF robotic manipulator was established by Lagrange method.The gain adaptive tuning iterative learning control method based on steady proportional derivative(PD)iterative learning control was proposed.The convergence of the designed controller was analyzed by using Lyapunov stability theory.Finally,in order to verify the effectiveness and superiority of the gain adaptive tuning iterative learning control method,a robotic manipulator control system was built in MATALB/Simulink for simulation analysis,and was compared with the steady PD iterative learning control method.The experimental results show that with the increase of the number of iterations,the tracking effect of the joint trajectory of the robotic manipulator is obviously improved.In the fifth iteration,the tracking of the joint angle trajectory and the end path of the robotic manipulator achieve the expected results.After the tenth iteration,the joint angular velocity trajectory of the robotic manipulator can also accurately track the desired trajectory.Compared with the steady PD iterative learning control,the gain adaptive tuning iterative learning control can achieve the desired trajectory with fewer iterations,faster learning speed and better trajectory tracking effect.This study provides an effective control method for a robotic manipulator with repetitive motion trajectory.
作者
李慧琴
颉世国
王硕飞
王洋洋
王鹏飞
兰明明
LI Huiqin;XIE Shiguo;WANG Shuofei;WANG Yangyang;WANG Pengfei;LAN Mingming(College of Mechanical and Electrical Engineering,Henan Agricultural University,Zhengzhou 450002,China)
出处
《机电工程》
北大核心
2025年第9期1810-1820,共11页
Journal of Mechanical & Electrical Engineering
基金
河南省科技研发计划联合基金资助项目(222103810026)。
