基于改进布谷鸟算法的机器人位置精度性能提升方法研究

Research on the Method of Improving Robot Positioning Accuracy Performance Based on Improved Cuckoo Algorithm

主要解决工业机器人在高端制造领域精度性能不足的问题。首先阐述了工业机器人误差模型的构建方法,将机器人运动学参数、关节减速比参数、耦合比参数进行统一建模;其次,提出了一种改进的布谷鸟搜索算法(Cuckoo Search Algorithm,CSA)的优化多参数辨识方法,利用对数调整系数修正Levy搜索步长,提升CSA优化算法的收敛性和精确性。为了验证提出的误差模型和优化方法的有效性,构建了串联型工业机器人标定实验系统。通过在Staubli TX60机器人的运动空间内测量160个测量点,分别构成辨识点集和验证点集。实验结果表明,待标定机器人的平均综合位置误差降低了86.7%以上。说明提出的误差模型和优化方法能够较好地提升工业机器人的精度性能。

It mainly focuses on the problem of insufficient accuracy performance of industrial robots in high-end manufacturing fields.Firstly,the construction method of industrial robot error model is presented.The kinematic parameters,joint reduction ratio parameters and coupling ratio parameters are uniformly modeled.Secondly,a multiple parameters identification method based on the improved cuckoo search algorithm(CSA)optimization algorithm is proposed.The convergence and accuracy of the CSA optimization algorithm are improved by apply-ing the logarithmic adjustment coefficient to correct the Levy search step.In order to verify the effectiveness of the proposed error model and optimization method,a serial industrial robot calibration experimental system is built.The identification point set and verification point set are composed of 160 measured points in the motion space of the Staubli TX60 robot.The average comprehensive position error of the calibrated robot is reduced by more than 86.7%.Therefore,it shows that the error model and optimization method proposed can better im-prove the accuracy performance of industrial robots.