多关节工业机器人精度增强方法研究

Research on Accuracy Enhancement Methods for Multi-Articulated Industrial Robots

中文关键词:  多关节机器人  工业机器人  误差标定  精度增强  运动学模型

英文关键词:multi-articulated robots  industrial robots  error calibration  accuracy enhancement  kinematic model

中文摘要:

      多关节工业机器人是当前工业应用中最为典型的机器人。为拓展多关节工业机器人在高精度领域的应用，对多关节工业机器人的精度增强方法开展了研究。首先，基于Denavit-Hartenberg(DH)模型构建了融合减速比参数的多关节工业机器人的运动学模型。其次，推导了多关节工业机器人的运动学误差模型。然后，搭建了多关节工业机器人的误差标定实验系统，其可实现精确的位姿误差测量。为提升运动学参数的辨识精度，在经典的正余弦算法（Sine-Cosine Algorithm,SCA）基础上提出了一种增加了个体最优值的个体迭代更新方法，保持了个体的多样性。实验结果表明，所提出的改进SCA能够将机器人的平均综合位置误差和平均综合姿态误差从（3.221 0 mm，0.027 25 rad）降低为（0.358 3 mm，0.017 13 rad），同时将机器人的最大综合位置误差和最大综合姿态误差从（8.859 6 mm，0.050 43 rad）降低为（1.209 6 mm，0.025 99 rad）。与经典SCA和粒子群优化（Particle Swarm Optimization,PSO）算法相比，所提出的改进SCA在位置精度的提升方面具有较大优势。

英文摘要:

      Multi-articulated industrial robots are the most typical robots in current industrial applications.To expand the application of multi-articulated industrial robots in the high-precision field,the research on the accuracy enhancement method of multi-articulated industrial robots is carried out.Firstly,the kinematic model of the multi-articulated industrial robot is constructed based on the Denavit-Hartenberg (DH) model by integrating the reduction ratio parameter.Secondly,the kinematic error model of the multi-articulated industrial robot is derived.Then,the error calibration experimental system of multi-articulated industrial robot is established,which can implement accurate position error measurement.In order to improve the identification accuracy of the kinematic parameters,the individual iterative update method of the individual optimal value is added to the classical sine-cosine algorithm (SCA),which can maintain the diversity of individuals.The experimental results show that the proposed improved SCA algorithm can reduce the average comprehensive position error and average comprehensive attitude error of the robot from (3.221 0 mm,0.027 25 rad) to (0.358 3 mm,0.017 13 rad).The maximum comprehensive position error and maximum comprehensive attitude error of the robot are also reduced from (8.859 6 mm,0.050 43 rad) to (1.209 6 mm,0.025 99 rad).Compared with the classical SCA and particle swarm optimization(PSO) algorithm,the proposed improved SCA has a greater advantage in position accuracy improvement.