Study on Multi-reservoir Echo State Network Modeling of Control Handing Comfort for Standing Posture
-
摘要: 操纵舒适性是人机系统研究的重要内容之一。针对操纵舒适性评价的不确定性和模糊性,构建基于贝叶斯的多核回声状态网络(Echo state network,ESN)模型,对站立姿态下的操纵舒适性进行评价。通过实验,收集不同操纵位置对用户的舒适性影响。15名被试者参与了本次实验,每个被试者需要完成100个操纵任务,实验过程中被试者的操纵姿势、目标位置、脚底压力、被试人体尺寸和主观舒适性的数据将被记录下来。选取了10%的实验数据对所提出的方法进行了验证,并与BP神经网络预测模型进行比较,结果表明T-S FNN模型具有较小的均方根误差。最后随机选取了10组操纵任务与快速上肢评估方法(RULA)进行比较,结果表明该方法和实际值相关性系数为0.97(p < 0.05),与RULA结果的相关性为0.66(p < 0.05),说明该方法能够较好地反应真实结果。Abstract: Control handing comfort is one of the most important study items of the man-machine system. Aiming at the uncertainty and fuzziness of control handing comfort evaluation, the aim of this study is to show the feasibility with a multi-reservoir echo state network (ESN) based on Bayesian regression to model comfort evaluation model for standing posture. The data for model training and testing were random selected. Fifteen adults participated in the experiment, they were asked to reach 100 different targets. Joint angle, foot pressure distribution, human characteristics, target position, and subjective comfort rating were collected during the experiment. The multi-reservoir ESN based on Bayesian regression was tested based on the 10% experimental data, the result showed that, the T-S FNN model has smaller root mean square error than BP neural network. In order to verify this model, 10 groups of different tasks were randomly selected, the results showed that the correlation coefficient between this method and the actual value was 0.97 (p < 0.05), with the rapid upper limb assessment (RULA) was 0.66(p < 0.05). It shows that this method can reflect the real results.
-
表 1 人体测量数据
性别 年龄/y 人数/位 体重/kg 身高/cm 肩高/cm 立姿举高/cm 侧面手臂长/cm 肩部指尖长/cm 坐高/cm 坐姿下肢长/cm 男 25.6±2.5 8 73.6±8.9 176.7±5.1 146.0±6.6 219.5±9.2 87.1±2.9 86.1±4.5 102.0±14.5 107.3±4.3 女 24.3±3.2 7 50.9±6.7 164.6±7.4 130.9±15.8 202.9±10.6 78.6±4.9 79.4±5.7 87.9±2.4 98.6±11.0 表 2 ESN与MrBESN的预测舒适性精度比较
模型 ESN MvBESN MvBESN MvBESN MvBESN 储备池数量 1 50 100 150 200 ERMS 8.51 4.83 4.26 3.87 3.95 表 3 模型评估结果
变量 操纵1 操纵2 操纵3 操纵4 操纵5 操纵6 操纵7 操纵8 操纵9 操纵10 目标位置 B1(2, 3) B3(3, 4) B4(1, 2) B2(3, 3) B3(1, 5) B2 (3, 3) B4(2, 2) B2(5, 1) B1(1, 4) B3(3, 4) 舒适性实际值 32 18 16 33 13 35 16 22 26 18 MvBESN得分 35.2 16.2 14.3 35.1 12.3 36.6 15.5 20.6 28.3 16.5 RULA得分 5 4 5 3 6 3 5 5 3 4 -
[1] Wang X G, Trasbot J. Effects of target location, stature and hand grip type on in-vehicle reach discomfort[J]. Ergonomics, 2011, 54(5):466-476 doi: 10.1080/00140139.2011.564312 [2] 苏胜, 王晗宇.基于JACK软件进行人机工程仿真分析的要点[J].工业设计, 2017, (11):117-118 doi: 10.3969/j.issn.1672-7053.2017.11.052Su S, Wang H Y. Core elements of ergonomics simulation analysis based on JACK[J]. Industrial Design, 2017, (11):117-118(in Chinese) doi: 10.3969/j.issn.1672-7053.2017.11.052 [3] Wirsching H J, Spitzhirn M. Virtual aging-implementation of age-related human performance factors in ergonomic vehicle design using the digital human model RAMSIS[C]//Proceedings of the 20th Congress of the International Ergonomics Association. Florence, Italy: Springer, 2019: 87-97 [4] Hanson L, Wienholt W, Sperling L. A control handling comfort model based on fuzzy logics[J]. International Journal of Industrial Ergonomics, 2003, 31(2):87-100 doi: 10.1016/S0169-8141(02)00177-4 [5] Golabchi A, Han S U, Fayek A R. A fuzzy logic approach to posture-based ergonomic analysis for field observation and assessment of construction manual operations[J]. Canadian Journal of Civil Engineering, 2016, 43(4):294-303 doi: 10.1139/cjce-2015-0143 [6] David G C. Ergonomic methods for assessing exposure to risk factors for work-related musculoskeletal disorders[J]. Occupational Medicine, 2005, 55(3):190-199 doi: 10.1093/occmed/kqi082 [7] Bao S, Howard N, Spielholz P, et al. Two posture analysis approaches and their application in a modified Rapid Upper Limb Assessment evaluation[J]. Ergonomics, 2007, 50(12):2118-2136 doi: 10.1080/00140130701458230 [8] Zenk R, Franz M, Bubb H, et al. Technical note: spine loading in automotive seating[J]. Applied Ergonomics, 2012, 43(2):290-295 doi: 10.1016/j.apergo.2011.06.004 [9] Chevalot N, Wang X G. An experimental investigation of the discomfort of arm reaching movements in a seated position[J]. SAE Transactions, 2004, 113(1):98-103 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=CC026940294 [10] Hulka K, Cuberek R, Svoboda Z. Time-motion analysis of basketball players: a reliability assessment of Video Manual Motion Tracker 1.0 software[J]. Journal of Sports Sciences, 2014, 32(1):53-59 doi: 10.1080/02640414.2013.805237 [11] 柴春雷.基于驾驶姿势预测模型的人机工程设计技术研究[D].杭州: 浙江大学, 2005Chai C L. Research on the technology of ergonomics design based on driving posture prediction model[D]. Hangzhou: Zhejiang University, 2005(in Chinese) [12] Jaeger H, Haas H. Harnessing nonlinearity: predicting chaotic systems and saving energy in wireless communication[J]. Science, 2004, 304(5667):78-80 doi: 10.1126/science.1091277 [13] 许美玲, 韩敏.多元混沌时间序列的因子回声状态网络预测模型[J].自动化学报, 2015, 41(5):1042-1046 http://d.old.wanfangdata.com.cn/Periodical/zdhxb201505018Xu M L, Han M. Factor echo state network for multivariate chaotic time series prediction[J]. Acta Automatica Sinica, 2015, 41(5):1042-1046(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/zdhxb201505018 [14] 韩敏, 穆大芸.基于贝叶斯回归的多核回声状态网络研究[J].控制与决策, 2010, 25(4):531-534, 541 http://d.old.wanfangdata.com.cn/Periodical/kzyjc201004010Han M, Mu D Y. Research on multi-reservoir ESN based on Bayesian regression[J]. Control and Decision, 2010, 25(4):531-534, 541(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/kzyjc201004010 [15] Tipping M E. Sparse bayesian learning and the relevance vector machine[J]. Journal of Machine Learning Research, 2001, 1(3):211-244 http://d.old.wanfangdata.com.cn/Periodical/dsjs201802008 [16] Hanson L, Wienholt W, Sperling L. A control handling comfort model based on fuzzy logics[J]. International Journal of Industrial Ergonomics, 2003, 31(2):87-100 doi: 10.1016/S0169-8141(02)00177-4 [17] Hulka K, Cuberek R, Svoboda Z. Time-motion analysis of basketball players: a reliability assessment of Video Manual Motion Tracker 1.0 software[J]. Journal of Sports Sciences, 2014, 32(1):53-59 doi: 10.1080/02640414.2013.805237 [18] Wininger M, Crane B A. Assessment of the minimally sufficient spatial sampling in pressure mapping the wheelchair seating interface[J]. Technology and Disability, 2015, 27(4):119-125 http://cn.bing.com/academic/profile?id=2b7a08a2a1fd0271dcaf101027dd6df2&encoded=0&v=paper_preview&mkt=zh-cn [19] McAtamney L, Nigel Corlett E. RULA: a survey method for the investigation of work-related upper limb disorders[J]. Applied Ergonomics, 1993, 24(2):91-99 doi: 10.1016/0003-6870(93)90080-S [20] Spearman秩相关系数界值表[EB/OL].(2014-06-26)[2018-12-31].https://wenku.baidu.com/view/ba8471aaaef8941ea76e05d0.html