基于GSO-RF意图识别算法的全身助力外骨骼控制方法研究

doi:10.3901/JME.2024.17.091

摘要/Abstract

摘要： 针对现有助力外骨骼结构复杂、功能单一的问题，基于人体工学理论，设计一套上下肢一体化的全身助力外骨骼，提升助力效果。针对助力外骨骼运动意图识别准确率不高的问题，提出一种基于位姿信号、肌电信号的运动意图识别方法，引入有限状态机(Finite state machine, FSM)，设定三种穿戴者运动状态；使用随机森林(Random forest, RF)算法对人机交互信号进行分类，确定穿戴者运动意图；采用人工萤火虫优化算法(Glowworm swarm optimization, GSO)优化随机森林，提高分类准确率，减少分类时间。为提高外骨骼控制系统轨迹跟踪精确性，保证外骨骼运动柔顺性，提出一种基于导纳自适应模糊反演算法的运动控制策略。搭建外骨骼实验平台，进行运动意图识别和轨迹跟踪实验，结果表明运动意图识别准确率可达96.6%，且实时性较好，提出的控制策略有效可行，轨迹跟踪效果较好。通过助力效能实验证明外骨骼的助力性能良好。

关键词: 全身助力外骨骼, 运动意图识别, 人机交互, 运动控制

Abstract: To address the complex structure and single function of existing power-assisted exoskeletons, a full-body power-assisted exoskeleton is designed based on the ergonomic theory, and the upper and lower limbs are integrated to improve the assisted effect. To address the problem of low accuracy of motion intention recognition of the power-assisted exoskeleton, a motion intention recognition method based on posture signals and electromyography signals is proposed. The random forest (RF) algorithm is used to classify the human-robot interaction signals and determine the wearer's movement intention; the Glowworm swarm optimization (GSO) algorithm is used to optimize the random forest to improve the classification accuracy and reduce the classification time. To improve the trajectory tracking accuracy of the exoskeleton control system and ensure the suppleness of the exoskeleton movement, a motion control strategy based on the guided adaptive fuzzy inversion algorithm is proposed. The exoskeleton experimental platform is built to carry out motion intention recognition and trajectory tracking experiments. The results show that the accuracy of motion intention recognition can reach 96.6%, and the real-time performance is good, the proposed control strategy is effective and feasible, and the trajectory tracking effect is good. The booster performance of the exoskeleton has been proven through booster effectiveness tests.

Key words: full-body power-assisted exoskeleton, motion intent recognition, human-robot interaction, motion control

中图分类号:

TG156

袁小庆, 吴涛, 原勋, 王文东. 基于GSO-RF意图识别算法的全身助力外骨骼控制方法研究[J]. 机械工程学报, 2024, 60(17): 91-101.

YUAN Xiaoqing, WU Tao, YUAN Xun, WANG Wendong. Research on a Full-body Power-assisted Exoskeleton Control Method Based on GSO-RF Intent Recognition Algorithm[J]. Journal of Mechanical Engineering, 2024, 60(17): 91-101.

参考文献

[1] AGUILAR-SIERRA H, YU W, SALAZAR S, et al. Design and control of hybrid actuation lower limb exoskeleton[J]. Advances in Mechanical Engineering, 2015, 7(6):1-13.
[2] 袁小庆, 姬俊杰, 刘宇轩, 等. 主被动结合的上下肢一体化助力外骨骼机器人的设计与效能评估[J]. 机械工程学报, 2022, 58(21):27-37. YUAN Xiaoqing, JI Junjie, LIU Yuxuan, et al. Design and performance evaluation of active-passive integrated exoskeleton robot with upper and lower limbs[J]. Journal of Mechanical Engineering, 2022, 58(21):27-37.
[3] LI W, CAO G, ZHU A. Review on control strategies for lower limb rehabilitation exoskeletons[J]. IEEE Access, 2021, 9:123040-123060.
[4] 李剑锋, 李国通, 张雷雨, 等. 穿戴式柔性下肢助力机器人发展现状及关键技术分析[J]. 自动化学报, 2020, 46(3):427-438. LI Jianfeng, LI Guotong, ZHANG Leiyu, et al. Advances and key techniques of soft wearable lower limb power-assisted robots[J]. Acta Automatica Sinica, 2020, 46(3):427-438.
[5] DE LOOZE M P, BOSCH T, KRAUSE F, et al. Exoskeletons for industrial application and their potential effects on physical work load[J]. Ergonomics, 2016, 59(5):671-681.
[6] 鲁守银, 袁鲁浩. 康复机器人的人机交互控制技术研究进展[J]. 山东建筑大学学报, 2021, 36(5):91-102. LU Shouyin, YUAN Luhao. Research progress of human-computer interaction control technology of rehabilitation robot[J]. Journal of Shandong Jianzhu University, 2021, 36(5):91-102.
[7] 张雷雨, 李剑锋, 刘钧辉, 等. 上肢康复外骨骼的设计与人机相容性分析[J]. 机械工程学报, 2018, 54(5):19-28. ZHANG Leiyu, LI Jianfeng, LIU Junhui, et al. Design and human-machine compatibility analysis of co-exos for upper-limb rehabilitation[J]. Journal of Mechanical Engineering, 2018, 54(5):19-28.
[8] SUTTER B, LELEVE A, PHAM M T, et al. A semi-autonomous mobile robot for bridge inspection[J]. Automation in Construction, 2018, 91:111-119.
[9] YUAN P, WANG T, MA F, et al. Key technologies and prospects of individual combat exoskeleton[C]//Knowledge Engineering and Management:Proceedings of the Seventh International Conference on Intelligent Systems and Knowledge Engineering, Beijing, China, Dec 2012(ISKE 2012). Springer Berlin Heidelberg, 2014:305-316.
[10] BOGUE R. Exoskeletons:a review of recent progress[J]. Industrial Robot, 2022, 49(5):813-818.
[11] CHRISTENSEN S, RAFIQUE S, BAI S. Design of a powered full-body exoskeleton for physical assistance of elderly people[J]. International Journal of Advanced Robotic Systems, 2021, 18(6):1-15.
[12] SATOH H, KAWABATA T, SANKAI Y. Bathing care assistance with robot suit HAL[C]//2009 IEEE International Conference on Robotics and Biomimetics (ROBIO). IEEE, 2009:498-503.
[13] LEE H, LEE B, KIM W, et al. Human-robot cooperation control based on a dynamic model of an upper limb exoskeleton for human power amplification[J]. Mechatronics, 2014, 24(2):168-176.
[14] LEE B, LEE H, LEE J, et al. Development of dynamic model-based controller for upper limb exoskeleton robot[C]//2012 IEEE International Conference on Robotics and Automation. IEEE, 2012:3173-3178.
[15] FONTANA M, VERTECHY R, MARCHESCH S, et al. The body extender:A full-body exoskeleton for the transport and handling of heavy loads[J]. IEEE Robotics & Automation Magazine, 2014, 21(4):34-44.
[16] 韩品刚. 基于多源信号融合的气动助力外骨骼人机协同控制研究[D]. 杭州:浙江大学, 2020. HAN Pingang. A study on human-robot cooperation control of pneumatic power-assisted exoskeleton based on multi-source information fusion[D]. Hangzhou:Zhejiang University, 2020.
[17] 屠尧, 朱爱斌, 宋纪元, 等. 下肢外骨骼康复机器人人机交互力自适应导纳控制[J]. 西安交通大学学报, 2019, 53(6):9-16. TU Yao, ZHU Aibin, SONG Jiyuan, et al. Adaptive admittance control of man-robot interaction force for lower limb exoskeleton rehabilitation robot[J]. Journal of Xi'an Jiaotong University, 2019, 53(6):9-16.
[18] 王文东, 李杰, 张俊博, 等. 基于迁移学习的手部离散动作识别方法研究[J]. 机械工程学报, 2022, 58(7):12-19. WANG Wendong, LI Jie, ZHANG Junbo, et al. Discrete hand motion recognition method based on transfer learning[J]. Journal of Mechanical Engineering, 2022, 58(7):12-19.
[19] 李海源, 刘畅, 严鲁涛, 等. 上肢外骨骼机器人的阻抗控制与关节试验研究[J]. 机械工程学报, 2020, 56(19):200-209. LI Haiyuan, LIU Chang, YAN Lutao, et al. Research on impendence control of an upper limb exoskeleton robot and joint experiments[J]. Journal of Mechanical Engineering, 2020, 56(19):200-209.
[20] 赵新刚, 谈晓伟, 张弼. 柔性下肢外骨骼机器人研究进展及关键技术分析[J]. 机器人, 2020, 42(3):365-384. ZHAO Xingang, TAN Xiaowei, ZHANG Bi. Development of soft lower extremity exoskeleton and its key technologies:a survey[J]. Robot, 2020, 42(3):365-384.
[21] GUNASEKARA J M P, GOPYRA R, JAYAWARDANE T S S, et al. Control methodologies for upper limb exoskeleton robots[C]//2012 IEEE/SICE International Symposium on System Integration (SII). IEEE, 2012:19-24.
[22] 王卫星. 面向运动意图识别的上肢外骨骼生物电信号控制研究[D]. 杭州:浙江大学, 2017. WANG Weixing. Research on bioelectric signal control of upper-limb exoskeleton for movement intention recognition[D]. Hangzhou:Zhejiang University, 2017.
[23] WAHYUNGGORO O, NUGROHO H A. Effect of window length on performance of the elbow-joint angle prediction based on electromyography[J]. Journal of Physics:Conference Series, 2017, 853(1):012014