协作SCARA机器人关节综合动态摩擦辨识与补偿

doi:10.3901/JME.2025.07.325

机械工程学报 ›› 2025, Vol. 61 ›› Issue (7): 325-337.doi: 10.3901/JME.2025.07.325

扫码分享

协作SCARA机器人关节综合动态摩擦辨识与补偿

汪朋朋¹, 卢浩², 杨志强¹, 侯福宁¹, 郭士杰^1,3, 甘中学¹

1. 复旦大学工程与应用技术研究院上海 200433;
2. 天津科技大学电子信息与自动化学院天津 300457;
3. 河北工业大学机械工程学院天津 300401

收稿日期:2024-04-04 修回日期:2024-08-15 发布日期:2025-05-12
作者简介:汪朋朋，男，1991年出生，博士研究生。主要研究方向为智能机器人，护理机器人。E-mail：19110860051@fudan.edu.cn
郭士杰(通信作者)，男，1963年出生，教授，博士研究生导师。主要研究方向为智能护理机器人、柔性步行助力机器人、无束缚生理信息检测、机器人智能传感技术。E-mail：guoshijie@fudan.edu.cn
基金资助:
上海科技计划(21511101701)和国家重点研发计划(2021YFC0122704)资助项目。

Comprehensive Dynamic Friction Identification and Compensation in Joints of Collaborative SCARA Robots

WANG Pengpeng¹, LU Hao², YANG Zhiqiang¹, HOU Funing¹, GUO Shijie^1,3, GAN Zhongxue¹

1. Academy for Engineering and Technology, Fudan University, Shanghai 200433;
2. College of Electronic Information and Automation, Tianjin University of Science and Technology, Tianjin 300457;
3. School of Mechanical Engineering, Hebei University of Technology, Tianjin 300401

Received:2024-04-04 Revised:2024-08-15 Published:2025-05-12

摘要/Abstract

摘要： 在协作机器人的动力学建模中，由于关节非线性摩擦力的存在，基于电机反馈电流的无力传感器力控精度不高。基于此，重点研究协作机器人关节非线性摩擦力，提高其辨识的精度，从而提升力矩控制性能。设计了一种新的协作SCARA机器人，对关节摩擦的组成进行了分析，并对影响摩擦的关节速度、温度和负载等因素进行了试验。通过采集机器人的位置、速度、力矩、环境温度和负载等时间序列数据，采用牛顿欧拉法计算理论力矩，将实际力矩作为监督学习的对象，建立基于Informer-LSTM的并联混合神经网络模型，从而辨识得出具有不同时间序列特征的非线性综合动态摩擦力，最终以前馈力矩的方式对关节摩擦进行补偿。实验结果表明，Informer-LSTM并联混合神经网络模型能够很好地预测长、短时间序列数据，以数据驱动的综合动态摩擦模型在零速和极低速下更为精确，摩擦补偿后的各关节力矩平均误差均在1%以下，能够显著提升力矩精度。

关键词: 协作SCARA机器人, 非线性摩擦力辨识, 摩擦补偿, 长短期记忆, Informer

Abstract: In the field of collaborative robot dynamics modeling, the presence of nonlinear friction forces at the joints results in low precision of force control based on motor feedback current. Against this backdrop, the research focuses on the nonlinear friction forces of collaborative robots to enhance the accuracy of their identification, thereby improving torque control performance. A new collaborative SCARA robot has been designed, analyzing the components of joint friction and conducting experiments on factors affecting friction, such as joint speed, temperature, and load. By collecting time series data of the robot’s position, speed, torque, ambient temperature, and load, the theoretical torque is calculated using the Newton-Euler method, the actual torque is used as the target for supervised learning to establish an Informer-LSTM-based parallel hybrid neural network model. This model identifies nonlinear composite dynamic friction forces with different time series characteristics, ultimately compensating for joint friction with feedforward torque. Experimental results show that the Informer-LSTM parallel hybrid neural network model can accurately predict long and short time series data. The data-driven comprehensive dynamic friction model is more precise at zero and very low speeds, with the average error in joint torque after friction compensation below 1%, significantly enhancing torque precision.

Key words: collaborative SCARA robot, nonlinear friction identification, friction compensation, long short-term memory, Informer

中图分类号:

TP242

汪朋朋, 卢浩, 杨志强, 侯福宁, 郭士杰, 甘中学. 协作SCARA机器人关节综合动态摩擦辨识与补偿[J]. 机械工程学报, 2025, 61(7): 325-337.

WANG Pengpeng, LU Hao, YANG Zhiqiang, HOU Funing, GUO Shijie, GAN Zhongxue. Comprehensive Dynamic Friction Identification and Compensation in Joints of Collaborative SCARA Robots[J]. Journal of Mechanical Engineering, 2025, 61(7): 325-337.

参考文献

[1] Gao L，Yuan J，Han Z，et al. A friction model with velocity，temperature and load torque effects for collaborative industrial robot joints[C]// 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). Vancouver，BC，Canada：IEEE，2017：3027-3032.
[2] Wolf S，Iskandar M. Extending a dynamic friction model with nonlinear viscous and thermal dependency for a motor and harmonic drive gear[C]// 2018 IEEE International Conference on Robotics and Automation (ICRA). Brisbane，Australia：IEEE，2018：783-790.
[3] Bittencourt A C，Axelsson P. Modeling and experiment design for identification of wear in a robot joint under load and temperature uncertainties based on friction data[J]. IEEE/ASME Transactions on Mechatronics，2013，19(5)：1694-1706.
[4] Huang S，Liang W，Tan K K. Intelligent friction compensation：A review[J]. IEEE/ASME Transactions on Mechatronics，2019，24(4)：1763-1774.
[5] 刘丽兰，刘宏昭，吴子英，等. 机械系统中摩擦模型的研究进展[J]. 力学进展，2008，38(2)：201-213. LIU Lilan，LIU Hongzhao，WU Ziying，et al. An overview of friction models in mechanical systems[J]. Advances in Mechanics，2008，38(2)：201-213.
[6] Armstrong-Helouvry B. Control of machines with friction[M]. Springer Science & Business Media，2012.
[7] Ravanbod-Shirazi L，BesanCon-Voda A. Friction identification using the Karnopp model，applied to an electropneumatic actuator[J]. Proceedings of the Institution of Mechanical Engineers，Part I：Journal of Systems and Control Engineering，2003，217(2)：123-138.
[8] Hess D P，Soom A. Friction at a lubricated line contact operating at oscillating sliding velocities[J]. Trans. ASME J. Tribology，1990，112(1)：147-152.
[9] Dahl P R. A solid friction model[J]. The Aerospace Corporation，1968，18(1)：1-24.
[10] De Wit C C，Olsson H，Astrom K J，et al. A new model for control of systems with friction[J]. IEEE Transactions on Automatic Control，1995，40(3)：419-425.
[11] Kikuuwe R，Takesue N，Sano A，et al. Fixed-step friction simulation：from classical Coulomb model to modern continuous models[C]// 2005 IEEE/RSJ International Conference on Intelligent Robots and Systems. Edmonton，Alberta，Canada：IEEE，2005：1009-1016.
[12] Xu L，Yao B. Adaptive robust control of mechanical systems with non-linear dynamic friction compensation[J]. International Journal of Control，2008，81(2)：167-176.
[13] Kabziński J，Jastrzębski M. Practical implementation of adaptive friction compensation based on partially identified LuGre model[C]// 201419th International Conference on Methods and Models in Automation and Robotics (MMAR). Miedzyzdroje，Poland：IEEE，2014：699-704.
[14] Tu X，Zhou Y F，Zhao P，et al. Modeling the static friction in a robot joint by genetically optimized BP neural network[J]. Journal of Intelligent & Robotic Systems，2019，94：29-41.
[15] Hirose N，Tajima R. Modeling of rolling friction by recurrent neural network using LSTM[C]// 2017 IEEE international conference on robotics and automation (ICRA). Singapore：IEEE，2017：6471-6478.
[16] Guo K，Pan Y，Yu H. Composite learning robot control with friction compensation：a neural network-based approach[J]. IEEE Transactions on Industrial Electronics，2018，66(10)：7841-7851.
[17] 李洋，朱立爽，刘今越，等. 基于动力学模型辨识的全臂柔顺控制[J]. 机械工程学报，2022，58(3)：45-54. LI Yang，ZHU Lishuang，LIU Jinyue，et al. Dynamic model identification for whole-arm compliance control[J]. Journal of Mechanical Engineering，2022，58(3)：45-54.
[18] 赵哲，欧屹，孟军虎，等. 考虑转速影响的滚珠丝杠副摩擦系数计算与试验研究[J]. 摩擦学学报，2017，37(6)：784-790. ZHAO Zhe，OU Yi，MENG Junhu，et al. Calculation and experimental study on friction coefficient of ball screws considering speed effect[J]. Tribology，2017，37(6)：784-790.
[19] 胡华君，潘博，孙京. 航天器驱动机构轴系摩擦力矩建模与分析研究[J]. 机械工程学报，2017，53(11)：75-80. HU Huajun，PAN Bo，SUN Jing. Friction torque modeling of spacecraft driving mechanism shafting[J]. Journal of Mechanical Engineering，2017，53(11)：75-80.
[20] Hochreiter S，Schmidhuber J. Long short-term memory[J]. Neural Computation，1997，9(8)：1735-1780.
[21] Brownlee J. Techniques to handle very long sequences with LSTMs [EB/OL]. [2019-08-14]. https：//machinelearningmastery.com/handle-long-sequences-long-short-term-memory-recurrent-neural-networks/.
[22] Bappy J H，Simons C，Nataraj L，et al. Hybrid lstm and encoder–decoder architecture for detection of image forgeries[J]. IEEE Transactions on Image Processing，2019，28(7)：3286-3300.
[23] Yang Z，Gjorgjevikj D，Long J，et al. Sparse autoencoder-based multi-head deep neural networks for machinery fault diagnostics with detection of novelties[J]. Chinese Journal of Mechanical Engineering，2021，34(1)：54.
[24] Pandiyan V，Akeddar M，Prost J，et al. Long short-term memory based semi-supervised encoder-decoder for early prediction of failures in self-lubricating bearings[J]. Friction，2023，11(1)：109-124.
[25] Vaswani A，Shazeer N，Parmar N，et al. Attention is all you need[J]. Advances in Neural Information Processing Systems，2017，30：5998-6008.
[26] Du Y，Zhang R，Shi P，et al. ST-LaneNet：Lane line detection method based on swin transformer and LaneNet[J]. Chinese Journal of Mechanical Engineering，2024，37(1)：14.
[27] Zeyer A，Bahar P，Irie K，et al. A comparison of transformer and lstm encoder decoder models for ASR[C]// 2019 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU). Singapore：IEEE，2019：8-15.
[28] Zhou H，Zhang S，Peng J，et al. Informer：Beyond efficient transformer for long sequence time-series forecasting[C]// Proceedings of the AAAI Conference on Artificial Intelligence. Virtually：AAAI，2021，35(12)：11106-11115.

协作SCARA机器人关节综合动态摩擦辨识与补偿

Comprehensive Dynamic Friction Identification and Compensation in Joints of Collaborative SCARA Robots

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 13

编辑推荐

Metrics

本文评价

[1]	刘占广, 张云, 刘晴雨. 基于差分融合长短期记忆神经网络数控机床热误差建模[J]. 机械工程学报, 2024, 60(7): 249-257.
[2]	林亮行, 马国政, 孙建芳, 韩翠红, 雍青松, 苏峰华, 王海斗. 基于VMD-EEMD-LSTM的涂层型关节轴承剩余使用寿命预测方法[J]. 机械工程学报, 2023, 59(9): 125-136.
[3]	李志兵, 肖健梅, 王锡淮. 基于多粒度NRS和改进Bi-LSTM的电力系统暂态稳定评估^*[J]. 电气工程学报, 2023, 18(3): 232-241.
[4]	李毅超, 王楠, 段彬, 康永哲, 张承慧. 基于深度学习和粒子群算法的锂离子电池核温估计方法[J]. 机械工程学报, 2023, 59(22): 69-78.
[5]	鲁南, 欧阳权, 黄俍卉, 王志胜. 基于注意力机制和多任务LSTM的锂电池容量预测方法^*[J]. 电气工程学报, 2022, 17(4): 41-50.
[6]	高凯, 李勋豪, 胡林, 陈彬, 杜荣华. 基于多头注意力的CNN-LSTM的换道意图预测[J]. 机械工程学报, 2022, 58(22): 369-378.
[7]	毛煜, 尚海昆, 于卓琦. 基于长短期记忆网络的电网同调机群快速辨识[J]. 电气工程学报, 2022, 17(2): 201-207.
[8]	蔡鑫祥, 撖奥洋, 周生奇, 菅学辉, 张智晟. 基于改进Bagging算法与模糊MP-LSTM融合的短期负荷预测模型^*[J]. 电气工程学报, 2022, 17(1): 164-170.
[9]	熊瑞, 闫良基, 王榘. 功率需求驱动的电动载运装备用动力电池充放电能力预测方法[J]. 机械工程学报, 2021, 57(20): 161-171.
[10]	余卓平, 韩伟, 熊璐, 徐松云. 基于Byrnes-Isidori标准型的集成式电子液压制动系统液压力控制^*[J]. 机械工程学报, 2016, 52(22): 92-100.
[11]	吴青聪, 王兴松, 杜峰坡, 陈林. 双套索系统力矩传动特性与摩擦补偿分析[J]. 机械工程学报, 2015, 51(5): 22-29.
[12]	吴跃飞;马大为;乐贵高. 基于修正LuGre模型的自适应鲁棒控制在机电伺服系统中的应用[J]. , 2014, 50(22): 207-212.
[13]	向红标;谭文斌;李醒飞;张晨阳. 基于LuGre模型的自适应摩擦补偿[J]. , 2012, 48(17): 70-74.