采用系统辨识技术识别水下航行器模型参数，是修正理论计算或船模水动力试验结果，实现精确数学建模的有效手段，对水下航行器总体性能优化、操纵性预报、控制器设计等方面都具有重要意义。重点对基于最小二乘法、人工智能和卡尔曼滤波的3种典型参数辨识方法进行分述，并对相关领域取得的成果、存在的问题和研究热点进行系统梳理。最后，对系统辨识技术在水下航行器模型参数辨识应用领域的发展趋势进行展望。
Identification of underwater vehicle model parameters by system identification technology is an effective way to correct the parameters obtained from theoretical calculation or ship model hydrodynamic test and realize accurate mathematical modeling, which is of great significance to overall performance optimization, maneuverability prediction and controller design for underwater vehicles. Three typical parameter identification methods based on Least Square, artificial intelligence and Kalman filter are described respectively, and the achievements, existing problems and research hotspots in related fields are systematically sorted out. Finally, the development trend of system identification technology in underwater vehicle model parameters identification is prospected.
2023,45(15): 81-86 收稿日期：2022-09-09
DOI：10.3404/j.issn.1672-7649.2023.15.015
分类号：U661
基金项目：装备预研重点实验室基金资助项目（6142217180201）
作者简介：吕帮俊(1981-),男,博士,副教授,研究方向为潜艇操纵和潜艇水动力
参考文献：
[1] 丁锋. 系统辨识(1): 辨识导引[J]. 南京信息工程大学学报, 2011, 3(1): 1–22
DING F. System identification. Part A: Introduction to the identification[J]. Journal of Nanjing University of Information Science & Technology, 2011, 3(1): 1–22
[2] 边靖伟, 寇立伟, 项基. 应用PSO和SVM的水下航行器黑箱建模[J]. 哈尔滨工业大学学报, 2019, 51(10): 55–60,82
BIAN J W, KOU L W, XIANG J. Black-box modeling based on PSO and SVM for underwater vehicles[J]. Journal of Harbin Institute of Technology, 2019, 51(10): 55–60,82
[3] YIN J, ZOU Z, XU F, et al. Online ship roll motion prediction based on grey sequential extreme learning machine[J]. Neurocomputing, 2014(129): 168–174
[4] 孙洪波, 施朝健. 基于 Elman 网络的船舶运动模型辨识[J]. 上海海事大学学报, 2014, 35(1): 10–13
[5] WANG X, ZOU Z, HOU X, et al. System identification modeling of ship manoeuvring motion based on support vector regression[J]. Journal of Hydrodynamics, 2015, 27(4): 502–512
[6] XU H, SOARES C. Vector field path following for surface marine vessel and parameter identification based on LS-SVM[J]. Ocean Engineering, 2016(113): 151–161
[7] 白伟伟, 任俊生, 李铁山, 等. 基于局部最优LWL的船舶操纵运动辨识建模[J]. 哈尔滨工程大学学报, 2017, 38(5): 676–683
[8] GERTLER M, HAGEN G R. Standard equations of motion for submarine simulation[R]. NSRDC Report 2510, June 1967.
[9] FIELDMAN, J. DTNSRDC revised standard submarine equations of motion[R]. Technical Re-port SPD-0393-09, David W. Taylor Naval Research and Development Center, USA.
[10] 施生达. 潜艇操纵性[M]. 北京: 国防工业出版社, 1995.
[11] 高婷, 庞永杰, 王亚兴, 等. 水下航行器水动力系数计算方法[J]. 哈尔滨工程大学学报, 2019, 40(1): 174–180
GAO T, PANG Y J, WANG Y X, et al. Calculation method of hydrodynamic coefficients for underwater vehicles[J]. Journal of Harbin Engineering University, 2019, 40(1): 174–180
[12] 邓志刚, 朱大奇, 方建安. 水下机器人动力学模型参数辨识方法综述[J]. 上海海事大学学报, 2014, 35(2): 74–80
[13] YOON H K, SON N S, LEE G J. Estimation of the Roll Hydrodynamic Moment Model of a Ship by Using the System Identification Method and the Free Running Model Test[J]. IEEE JOURNAL OF OCEANIC ENGINEERING, 2007, 32(4): 798–806
[14] FURLONG M, HEARN G E, VERES S M, et al. Nonlinear system identification tools applied to the modeling of submarine dynamics[C]//1st IFAC Workshop on Guidance and Control of Underwater Vehicles (GCUV 2003).
[15] SABET M T, DANIALI H M, SABET A F. Identification of an autonomous underwater vehicle hydrodynamic model using the extended, cubature, and transformed unscented Kalman filter[J]. IEEE journal of oceanic engineering, 2018, 43(2): 457–467
[16] KIM D. Estimation of hydrodynamic coefficients from results of real ship sea trials[J]. Polish Maritime Research, 2018, 25: 65–72
[17] 孙明轩, 毕宏博. 学习辨识: 最小二乘算法及其重复一致性[J]. 自动化学报, 2012, 38(5): 698–706
[18] 龚涛, 董早鹏. 基于RLSM 的海洋无人航行器操纵性参数辨识[J]. 大连海事大学学报, 2019, 45(2): 50–57
[19] 姜晓政, 刘 忠, 张建强, 等. 无人水面艇响应模型的参数辨识方法研究[J]. 计算机与数字工程, 2019, 47(1): 136–139,151
[20] 朱红坤, 郭蕴华, 牟军敏, 等. 基于多传感器递推总体最小二乘融合的水下机器人动力学模型参数辨识[J]. 船舶力学, 2017, 21(10): 1263–1270
ZHU H K, GUO Y H, MOU J M, et al. Dynamics model identification of underwater vehicles based on the multi-sensor fusion of recursive total least squares[J]. Journal of Ship Mechanics, 2017, 21(10): 1263–1270
[21] 林子淇. 水下机器人动力学建模与系统辨识技术研究[D]. 哈尔滨: 哈尔滨工程大学, 2015.
[22] 秦余钢, 马勇, 张亮, 等. 基于改进最小二乘算法的船舶操纵性参数辨识[J]. 吉林大学学报(工学版), 2016, 46(3): 897–903
[23] 于华男. 开架式水下机器人辨识与控制技术研究[D]. 哈尔滨: 哈尔滨工程大学, 2003.
[24] 丁锋. 系统辨识(6): 多新息辨识理论与方法[J]. 南京信息工程大学学报, 2012, 4(1): 1–28
[25] 朱胜庭, 朱大奇, 邓志刚. 多新息最小二乘法辨识水下机器人动力学模型[J]. 系统仿真学报, 2013, 25(6): 1399–1404
[26] 谢朔, 初秀民, 柳晨光, 等. 基于多新息最小二乘法的船舶操纵响应模型参数辨识[J]. 中国航海, 2017, 4(1): 73–78
[27] 陈玮琪, 颜 开, 王宝寿. 辨识航行体水动力参数的智能方法[J]. 船舶力学, 2011, 5(4): 359–363
CHEN W Q, YAN K, WANG B S. Intelligent identification of Hydrodynamic parameters of navigating body[J]. Journal of Ship Mechanics, 2011, 5(4): 359–363
[28] 陈玮琪, 颜开, 史淦君, 等. 水下航行体水动力参数智能辨识方法研究[J]. 船舶力学, 2007, 11(1): 40–46
CHEN W Q, YAN K, SHI G J, et al. Research of hydrodynamic parameter identification for underwater vehicle using swarm intelligence algorithm[J]. Journal of Ship Mechanics, 2007, 11(1): 40–46
[29] 何斌, 万磊, 姜大鹏, 等. 基于预测模型的模糊参数自寻优S面控制器[J]. 哈尔滨工程大学学报, 2014, 35(3): 267–273
HE B, WAN L, JIANG D P, et al. Fuzzy parametre sefl-optimized S surface controller based on the prediction model[J]. Journal of Harbin Engineering University, 2014, 35(3): 267–273
[30] 张剑. 基于神经网络的水下机器人建模方法研究[D]. 哈尔滨: 哈尔滨工程大学, 2006.
[31] 蒋 帆, 徐海祥, 余文瞾, 等. 基于自适应重组遗传算法的动力定位船舶水动力参数辨识[J]. 武汉理工大学学报(交通科学与工程版), 2019, 43(2): 276–280
[32] 袁伟杰. 自治水下机器人动力学建模及参数辨识研究[D]. 青岛: 中国海洋大学, 2010.
[33] 李庆梅, 刘平, 潘芳煜, 等. 基于NSGA-II算法的水下机器人水平面运动参数辨识[J]. 上海第二工业大学学报, 2016, 33(2): 141–145
[34] 朱大奇, 袁义丽, 邓志刚. 水下机器人参数辨识的量子粒子群算法[J]. 控制工程, 2015, 22(1): 531–537
[35] 田延飞, 黄立文, 熊勇, 等. 利用新型蝙蝠算法辨识船舶运动模型参数[J]. 科学技术与工程, 2018, 18(8): 136–143
TIAN Y F, HUANG L W, XIONG Y, et al. Parameter identification of ship motion model by using novel bat algorithm[J]. Science Technology and Engineering, 2018, 18(8): 136–143
[36] 田延飞, 胡山川, 黄立文, 等. 基于人工蜂群算法的船舶运动模型参数离线辨识[J]. 信息技术, 2017(11): 49–52
[37] 王建国, 姜春萌, 孙玉山, 等. 基于GA和BP混合算法的水下机器人系统建模[J]. 中南大学学报(自然科学版), 2011, 42(S1): 427–431
WANG J G, JIANG C M, SUN Y S, et al. Neural network identification of underwater vehicle by hybrid learning algorithm[J]. Journal of Central South University (Science and Technology), 2011, 42(S1): 427–431
[38] 周军, 王字飞. 基于PSO-BP算法的水下机器人运动模型辨识[J]. 机械与电子, 2013(3): 66–69
[39] 夏韬, 周军. 基于CPSO-BP算法的水下机器人神经网络辨识[J]. 机械与电子, 2013(10): 71–74
[40] 徐锋. 基于支持向量机的水下航行器动力学辨识建模研究[D]. 上海: 上海交通大学, 2013.
[41] 谢朔, 初秀民, 柳晨光, 等. 基于改进LSSVM的船舶操纵运动模型在线参数辨识方法[J]. 中国造船, 2018, 59(2): 178–189
XIE S, CHU X M, LIU C G, et al. Online parameter identification method for ship maneuvering models based on improved LSSVM[J]. Shipbuilding of China, 2018, 59(2): 178–189
[42] 张心光, 邹早建, 王岩松. 基于支持向量回归机和粒子群算法的船舶操纵运动模型辨识[J]. 船舶力学, 2016, 20(11): 1427–1432
ZHANG X G, ZOU Z J, WANG Y S. Identification of models of ship manoeuvring motion using Support vector regression and particle swarm optimization[J]. Journal of Ship Mechanics, 2016, 20(11): 1427–1432
[43] KALMAN R E. A new approach to linear filtering and prediction problems[J]. Journal of Basic Engineering 1960, 82 (1): 35–45.
[44] ERNANI M Z, BOZORG M, EBRAHIMI S. Identification of an autonomous underwater vehicle dynamic using extended kalman filter with ARMA noise model[J]. International Journal of Robotics, 2015, 4(1): 22–28
[45] 吕帮俊, 黄斌, 明廷涛, 等. 基于卡尔曼滤波的水下航行器水动力系数辨识方法[J]. 武汉理工大学学报(交通科学与工程版), 2021, 45(3): 463–469
LV B J, HUANG B, MING T T, et al. Hydrodynamic coefficients identification method for underwater vehicles based on Kalman filter[J]. Journal of Wuhan University of Technology (Transportation Science & Engineering), 2021, 45(3): 463–469
[46] ABKOWITZ M A. Measurement of hydrodynamic characteristics from ship maneuvering trials by system identification[J]. SNAME transaction, 1980, 33(88): 283–318
[47] SHARIATI H, MOOSAVI H, DANESH M. Application of particle filter combined with extended Kalman filter in model identification of an autonomous underwater vehicle based on experimental data[J]. Applied Ocean Research, 2019, 82: 32–40
[48] 赵大明, 施朝健, 彭静. 应用扩展卡尔曼滤波算法的船舶运动模型参数辨识[J]. 上海海事大学学报, 2008, 29(3): 5–9
[49] 丁彦侃, 俞孟蕻. 并行扩展卡尔曼滤波的船舶模型参辨识研究[J]. 船舶工程, 2015, 37(1): 72–74,99
[50] CARDENAS P, DEBARROS E A. Estimation of AUV hydrodynamic coefficients using analytical and system identification approaches[J]. IEEE Journal of Oceanic Engineering, 2019, 0364-9059;1558-1691.
[51] JULIER S J, UHLMANN J K, DURRANT-Whyte H F. A new approach for filtering nonlinear systems [C]// Proceeds of American Control Conference, Seattle, WA, USA: IEEE, 1995, 3: 1628―1632.
[52] JULIER S J, UHLMANN J K. A new extension of the Kalman filter to nonlinear systems [C]// Proceeds of the 11th International Symposium on Aerospace/Defense Sensing, Simulation and Controls. Society of Photo-optical Instrumentation Engineers, Orlando, Florida, USA: 1997: 182-193.
[53] SABET M T, SARHADI P, ZARINI M. Extended and unscented kalman filters for parameter estimation of an autonomous underwater vehicle[J]. Ocean Engineering, 2014, 91: 329–339
[54] van der Merwe R, WAN E A. The square root unscented Kalman filter for state and parameter estimation[C]//Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing. New York, 2001: 3461–3464.
[55] BELANGER F, MILLAN D, ENG P, et al. Submarine autopilot performance optimization with system identification[C]//Proceedings of the International Ship Control Systems Symposium (iSCSS), 2018.10.
