基于ISSA-LSSVR的6-DoF机器人定位精度补偿研究

doi:10.12068/j.issn.1005-3026.2025.20240136

摘要/Abstract

摘要：

为提高6自由度(6-DoF)机器人的定位精度，提出一种6-DoF机器人定位误差预测和精度补偿方法.介绍了在机器人的高频工作区内的分层逐行采样方法，并建立累积测量误差修正公式提高测量的准确性.实测证明机器人工作位置直接影响绝对误差.为此，提出了基于改进麻雀搜索算法优化最小二乘支持向量回归(ISSA-LSSVR)算法的误差补偿模型，预测和修正机器人自身定位误差.结果表明，相较于支持向量回归(SVR)、最小二乘支持向量回归(LSSVR)和麻雀搜索算法优化最小二乘支持向量回归(SSA-LSSVR)算法，基于ISSA-LSSVR算法的误差补偿效果最好，机器人的绝对误差降低了65.68%，最大误差降低了68.95%.

关键词: 6自由度机器人, 误差测量, 累积测量误差, ISSA-LSSVR算法, 误差补偿

Abstract:

To enhance the positioning accuracy of six-degree-of-freedom （6-DoF） robots， a method for predicting and compensating for positioning errors of the 6-DoF robot was proposed. A hierarchical line-by-line sampling strategy in the high-frequency workspace of the robot was introduced， and a cumulative measurement error correction formula was established to improve measurement accuracy. Experimental results demonstrate that the robot’s working position significantly affects absolute errors. To address this issue， an error compensation model based on an improved sparrow search algorithm-optimized least squares support vector regression （ISSA-LSSVR） was developed to predict and correct the robot’s inherent positioning errors. The results indicate that， relative to the support vector regression （SVR） algorithm， least squares support vector regression （LSSVR） algorithm， and sparrow search algorithm-optimized LSSVR （SSA-LSSVR）， the ISSA-LSSVR algorithm achieves superior compensation performance. Specifically， the absolute error is reduced by 65.68%， and the maximum error is decreased by 68.95%.

Key words: 6-DoF robot, error measurement, cumulative measurement error, ISSA-LSSVR algorithm, error compensation

中图分类号:

TP 242.2

于华宇, 朱文福, 辛博, 孙俊峰. 基于ISSA-LSSVR的6-DoF机器人定位精度补偿研究[J]. 东北大学学报（自然科学版）, 2025, 46(12): 48-56.

Hua-yu YU, Wen-fu ZHU, Bo XIN, Jun-feng SUN. Research on Positioning Accuracy Compensation of 6-DoF Robots Based on ISSA-LSSVR[J]. Journal of Northeastern University(Natural Science), 2025, 46(12): 48-56.

图/表 13

图1 机器人误差测量系统

Fig.1 Robot error measurement system

图2 采样点规划

Fig.2 Sampling point planning

图3 理论采样点与实测点的轨迹对比

Fig.3 Comparison of trajectory between theoretical sampling points and measured points

图4 测量误差的产生和累积

Fig.4 Generation and accumulation of measurement

图5 测量点样本的绝对误差

Fig.5 Absolute error of measured point samples

图6 第4、第6层测量点样本的绝对误差（a）—第4层测量点样本；（b）—第6层测量点样本.

Fig.6 Absolute error of measured point samples in the 4th and 6th layers

图7 误差概率密度分布拟合结果（a）—X轴方向误差正态分布拟合；（b）—Y轴方向误差正态分布拟合；（c）—Z轴方向误差正态分布拟合.

Fig.7 Distribution fitting results of error probability density

图8 LSSVR算法结构

Fig.8 LSSVR algorithm structure

图9 实验点位置测量

Fig.9 Experimental point position measurement

图10 实验点各方向误差预测结果

Fig.10 Experimental point error prediction results in all directions

图11 机器人X轴误差训练集预测精度对比（a）—SVR算法；（b）—LSSVR算法；（c）—SSA-LSSVR算法；（d）—ISSA-LSSVR算法.

Fig.11 Comparison of prediction accuracy of robot’s X-axis error training set

图12 补偿后误差对比

Fig.12 Error comparison after compensation

表1 不同SVR算法的补偿效果

Table 1 Compensation effect of different SVR algorithms

算法	绝对误差最大值/mm	绝对误差平均值/mm	方差/mm²
未补偿	0.380 4	0.169 4	0.005 4
SVR	0.207 0	0.080 9	0.001 8
LSSVR	0.164 8	0.069 3	0.001 1
SSA-LSSVR	0.133 1	0.059 4	0.000 7
ISSA-LSSVR	0.118 4	0.057 5	0.000 6

参考文献 15

[1]	Peta K， Wlodarczyk J， Maniak M. Analysis of trajectory and motion parameters of an industrial robot cooperating with a numerically controlled machine tools［J］.Journal of Manufacturing Processes，2023，101：1332-1342.
[2]	Liu Y， Wang D， Mi J H， et al. Advances in reliability and maintainability methods and engineering applications： essays in honor of professor Hong-zhong Huang on his 60th birthday［M］. Cham： Springer，2023.
[3]	Zeng Y F， Tian W， Liao W H. Positional error similarity analysis for error compensation of industrial robots［J］.Robotics and Computer Integrated Manufacturing，2016，42：113-120.
[4]	Li R， Ding N， Zhao Y， et al. Real-time trajectory position error compensation technology of industrial robot［J］.Measurement，2023，208：112418.
[5]	Wang Z， Zhang R N， Keogh P. Real-time laser tracker compensation of robotic drilling and machining［J］.Journal of Manufacturing and Materials Processing，2020，4（3）：79.
[6]	Yang B， Yang B N， Liu J D. Research on adjustable baseline binocular vision measurement system［J］.International Journal of Frontiers in Engineering Technology，2022，4（7）：30-34.
[7]	Nubiola A， Bonev I A. Absolute calibration of an ABB IRB 1600 robot using a laser tracker［J］.Robotics and Computer-Integrated Manufacturing，2013，29（1）： 236-245.
[8]	Gao G B， Liu F， San H J， et al. Hybrid optimal kinematic parameter identification for an industrial robot based on BPNN-PSO［J］.Complexity，2018，2018（1）：4258676.
[9]	王龙飞，李旭，张丽艳，等.工业机器人定位误差规律分析及基于ELM算法的精度补偿研究［J］.机器人，2018，40（6）：843-851，859.
	Wang Long-fei， Li Xu， Zhang Li-yan， et al. Analysis of the positioning error of industrial robots and accuracy compensation based on ELM algorithm［J］. Robot，2018，40（6）：843-851，859.
[10]	Ryu D， Choi O， Baik J. Value-cognitive boosting with a support vector machine for cross-project defect prediction［J］.Empirical Software Engineering，2016，21（1）：43-71.
[11]	薛祥儒，张承瑞，胡天亮，等.基于PSO-SVR算法的工业机器人分级标定方法［J］.计算机集成制造系统，2023，29（1）：51-60.
	Xue Xiang-ru， Zhang Cheng-rui， Hu Tian-liang， et al. Hierarchical calibration method of industrial robots based on PSO-SVR algorithm［J］. Computer Integrated Manufacturing Systems， 2023，29（1）：51-60.
[12]	Liu Z Y， Xu Y K， Duan G F， et al. Accurate on-line support vector regression incorporated with compensated prior knowledge［J］.Neural Computing and Applications，2021，33（15）：9005-9023.
[13]	Yu L A， Xu H J， Tang L. LSSVR ensemble learning with uncertain parameters for crude oil price forecasting［J］.Applied Soft Computing，2017，56：692-701.
[14]	Thomas S， Pillai G N， Pal K. Prediction of peak ground acceleration using ε -SVR， ν-SVR and Ls-SVR algorithm［J］.Geomatics， Natural Hazards and Risk，2017，8（2）：177-193.
[15]	Xue J K， Shen B. A novel swarm intelligence optimization approach： sparrow search algorithm［J］.Systems Science & Control Engineering，2020，8（1）：22-34.