摘 要:
為滿足欠驅(qū)動自主水下航行器(autonomous underwater vehicle, AUV)在復(fù)雜擾動和參數(shù)不確定條件下高性能軌跡跟蹤需求,提出預(yù)設(shè)動態(tài)性能及收斂時間的三維軌跡跟蹤控制方法。首先,對欠驅(qū)動AUV的前向位置道進行擴維,構(gòu)建面向控制的一體化多輸入多輸出軌跡跟蹤模型。然后,結(jié)合動態(tài)過程函數(shù)與預(yù)設(shè)時間控制理論,建立動態(tài)性能預(yù)設(shè)軌跡跟蹤控制系統(tǒng),使得AUV軌跡跟蹤暫態(tài)品質(zhì)可由動態(tài)過程函數(shù)直接決定,而跟蹤誤差的實際收斂時間也可由單個控制參數(shù)準確預(yù)設(shè)。最后,為避免控制奇異現(xiàn)象和“微分爆炸”現(xiàn)象,控制系統(tǒng)設(shè)計過程中分別融入絕對值修正法和徑向基函數(shù)網(wǎng)絡(luò)(radial basis function neural network, RBFNN)擬合法。數(shù)值仿真結(jié)果表明,所提出的控制方法可顯著提升欠驅(qū)動AUV的抗擾性和暫態(tài)品質(zhì),實現(xiàn)快速平滑的高性能三維軌跡跟蹤。
關(guān)鍵詞:
自主式水下航行器; 動態(tài)過程函數(shù); 預(yù)設(shè)時間控制理論; 動態(tài)性能預(yù)設(shè)軌跡跟蹤控制; 徑向基函數(shù)網(wǎng)絡(luò)
中圖分類號:
U 66
文獻標志碼: A""" DOI:10.12305/j.issn.1001-506X.2024.09.30
Trajectory tracking control with predefined dynamic performance for
underactuated autonomous underwater vehicle
LI Xiaobin*, XU Dong, YANG Xue
(Unit 92942 of the PLA, Beijing 100071, China)
Abstract:
To meet the high-performance trajectory tracking requirements of underactuated autonomous underwater vehicle (AUV) under complex disturbances and parameter uncertainty, a three-dimensional trajectory tracking controller method with predefined dynamic performance and convergence time is proposed. Firstly, by extending the forward position channel of the underactuated AUV, and a multi-input multi-output trajectory tracking control-oriented model is developed. Subsequently, by combining dynamic process functions with predefined-time control theory, a dynamic performance-predefined trajectory tracking control system is established, which allows the transient quality of AUV trajectory tracking to be determined by dynamic process functions, and the actual convergence time of tracking errors to be predefined by a single control parameter. Finally, to avoid control singularities and the “differential explosion” phenomenon, the controller design incorporates the absolute value correction method and radial basis function neural network (RBFNN) fitting method. Numerical simulation results indicate that the proposed controller significantly improves the disturbance rejection and transient quality of underactuated AUV, achieving fast, smooth, and high-performance trajectory tracking.
Keywords:
autonomous underwater vehicle (AUV); dynamic process function; predefined-time control theory; dynamic performance-predefined trajectory tracking control; radial basis function neural network (RBFNN)
0 引 言
自主水下航行器(autonomous underwater vehicle, AUV)以其便捷性、快速性和經(jīng)濟性,在石油和天然氣勘探、深海檢測、海洋測繪、管道維護和軍事應(yīng)用等任務(wù)中重要性日益增加[1]。精確控制AUV運動,實現(xiàn)高性能路徑跟蹤[2-3]或軌跡跟蹤[4-6]對高效完成多樣化任務(wù)尤為關(guān)鍵。相較于路徑跟蹤,軌跡跟蹤要求控制律導(dǎo)引AUV跟蹤具有時變特性的參考軌跡,應(yīng)用范圍更廣、更具挑戰(zhàn)性[4],具體表現(xiàn)在:① 為適應(yīng)時變參考軌跡,軌跡跟蹤誤差需在有限時間區(qū)間內(nèi)收斂[5];② 考慮成本、總重量和效率,AUV實際運動控制執(zhí)行器通常為欠驅(qū)動配置[6];③ AUV動力學(xué)模型呈現(xiàn)高度非線性、強耦合、參數(shù)不確定特性,且受未知時變外部擾動影響[6]。因此,面向AUV軌跡跟蹤需求,應(yīng)研究具有強魯棒、抗擾性和良好動態(tài)性能,且適應(yīng)欠驅(qū)動性和收斂時間約束的控制方法。
現(xiàn)有復(fù)雜擾動下AUV模型的常用方法包括反步控制[7-8]、滑模控制[9-10]、神經(jīng)網(wǎng)絡(luò)控制[11-12]、模糊控制[13-14]等。Wu等[7]和周鑄等[8]提出反步抗擾控制方法實現(xiàn)AUV軌跡跟蹤誤差的最終一致有界收斂。李娟等[9]和李鑫濱等[10]利用滑模方法的強魯棒性實現(xiàn)了AUV軌跡跟蹤誤差的漸近收斂。神經(jīng)網(wǎng)絡(luò)和模糊系統(tǒng)以其對連續(xù)有界擾動的有效逼近特性,成為增強抗擾性的重要控制工具。文獻[11-12]利用徑向基函數(shù)神經(jīng)網(wǎng)絡(luò)(radial basis function neural network, RBFNN)自適應(yīng)估計與補償外部擾動,實現(xiàn)軌跡跟蹤誤差的漸近收斂。劉用等[13]設(shè)計AUV縱向平面與水平面運動穩(wěn)定模糊控制器,Liang等[14]則設(shè)計了自適應(yīng)模糊動態(tài)面控制方法。雖然文獻[7-14]均設(shè)計了滿足各自目標的控制系統(tǒng),但僅有文獻[8,11,14]研究了欠驅(qū)動AUV的三維軌跡跟蹤問題,所實現(xiàn)的軌跡跟蹤誤差漸近或有限時間收斂難以匹配快速收斂需求。同時,盡管文獻[11]使用障礙Lyapunov函數(shù)與預(yù)設(shè)性能函數(shù)實現(xiàn)AUV軌跡跟蹤的期望動態(tài)性能,但在實際使用時均存在狀態(tài)量超出預(yù)定包絡(luò)而導(dǎo)致控制失穩(wěn)的風(fēng)險。
為加快跟蹤誤差收斂速率,部分研究將固定時間穩(wěn)定性引入AUV軌跡跟蹤控制系統(tǒng)中[15-21]。Chen等[15]設(shè)計自適應(yīng)固定時間控制律實現(xiàn)跟蹤誤差在固定時間內(nèi)收斂至原點附近鄰域。Zheng等[16]和Sun等[17]均提出基于固定時間擴張狀態(tài)觀測器的AUV控制器,使得跟蹤誤差可在未知狀態(tài)和外部集總擾動下固定時間收斂至原點。類似的固定時間控制律還呈現(xiàn)在文獻[18-21]中。盡管該類方法可使預(yù)估收斂時間且與初始狀態(tài)無關(guān),但控制參數(shù)與預(yù)估收斂時間關(guān)系復(fù)雜,且實際收斂時間遠小于預(yù)估收斂時間,從而使得初始收斂速度過大,易出現(xiàn)執(zhí)行機構(gòu)飽和及振蕩。
為避免固定時間控制參數(shù)與預(yù)估收斂時間復(fù)雜關(guān)系,單參數(shù)決定收斂時間的預(yù)設(shè)時間控制理論[22]逐步興起,并已被應(yīng)用于AUV軌跡跟蹤控制。Sun等[23]基于預(yù)設(shè)時間性能函數(shù)和反步法設(shè)計了面向動態(tài)性能需求的全驅(qū)AUV軌跡跟蹤控制器。Li等[24]設(shè)計了預(yù)設(shè)時間滑??刂破骱陀^測器,使得全驅(qū)AUV的軌跡跟蹤誤差在預(yù)設(shè)時間內(nèi)收斂至原點附近鄰域。Li等[25]基于預(yù)設(shè)時間Lyapuynov范式動力學(xué)建立自適應(yīng)預(yù)設(shè)時間最優(yōu)軌跡跟蹤控制器。盡管在文獻[24-25]中預(yù)估收斂時間由單個參數(shù)決定,但未有效減少預(yù)估保守性。而文獻[23]中基于預(yù)設(shè)時間性能函數(shù)的控制方法在有限采樣頻率下存在控制系統(tǒng)失效風(fēng)險,工程實用性尚待提升。目前預(yù)設(shè)時間控制理論研究從機理上可分為3類。① 預(yù)設(shè)時間邊界函數(shù)方法:設(shè)定預(yù)設(shè)時間收斂的邊界函數(shù),并保證系統(tǒng)狀態(tài)量始終在邊界內(nèi),但存在控制失效風(fēng)險[26-27];② 時不變Lyapunov動力學(xué)范式方法:預(yù)估保守性大,初期控制量易飽和[28-29];③ 時變Lyapunov動力學(xué)范式方法:基于趨向于無窮的時變函數(shù),在有限采樣頻率下實用性受限[30-31]。因此,需設(shè)計一種具有工程實用的新型預(yù)設(shè)時間控制系統(tǒng),以滿足AUV軌跡跟蹤控制的快速性與動態(tài)性能需求。
綜合調(diào)研分析,本文針對欠驅(qū)動AUV三維軌跡跟蹤控制問題,設(shè)計預(yù)設(shè)時間及動態(tài)性能的三維軌跡跟蹤控制系統(tǒng),主要貢獻包括:① 基于擴維一體化輸入輸出模型設(shè)計欠驅(qū)動AUV自適應(yīng)軌跡跟蹤控制器,避免繁瑣的分通道設(shè)計流程;② 未知擾動和參數(shù)不確定條件下,單個控制參數(shù)即可準確設(shè)定軌跡跟蹤誤差實際收斂時間,有效降低收斂過程的控制量需求;③ 通過引入動態(tài)過程函數(shù),使得軌跡跟蹤誤差動態(tài)性能簡便可調(diào),避免大跟蹤誤差條件下模型狀態(tài)量振蕩現(xiàn)象,具有較好工程實用性。
全文組織如下:第1節(jié)給出常用引理、定理,推導(dǎo)面向控制的欠驅(qū)動AUV模型,并建立控制問題;第2節(jié)完成預(yù)設(shè)時間控制系統(tǒng)推導(dǎo)和全系統(tǒng)穩(wěn)定性論證;第3節(jié)通過數(shù)值仿真驗證所提方法在參數(shù)不確定和外部擾動下的高控制性能;最后,在第4節(jié)中給出總結(jié)。
3.2 變時間約束仿真分析
考慮AUV軌跡跟蹤誤差快速收斂需求,針對不同的收斂時間要求,設(shè)定情況1:TAUV=18 s;情況2:TAUV=20 s;情況3:TAUV=25 s。為凸顯軌跡跟蹤的動態(tài)品質(zhì),對前30 s航行過程展開仿真,結(jié)果如圖4~圖9所示。
分析圖4~圖7可知,在本文所提預(yù)設(shè)時間控制方法的作用下,AUV在3種情況中均實現(xiàn)了對標稱軌跡的高精度跟蹤,且跟蹤誤差的實際收斂時間與期望收斂時間相同,顯示出本文方法對存在時間約束的AUV軌跡跟蹤問題具有良好的適用性。
而由圖8和圖9可知,航行全程俯仰角和偏航角變化平滑,除去初始階段和收斂終端階段姿態(tài)角較大幅變化外,其他階段姿態(tài)角保持小值,未出現(xiàn)振蕩現(xiàn)象。
3.3 對比仿真分析
本節(jié)將引入文獻[14]中控制方法作為對比方法1。同時,為驗證本文方法中自適應(yīng)環(huán)節(jié)的作用,將控制律中自適應(yīng)項移除作為對比方法2。對比方法1的參數(shù)設(shè)置為:k1=0.3,k2=1.5,k3=0.5,2=0.2,ρ1=0.5,ρ2=0.8,ρ3=2。對全程軌跡展開擾動和參數(shù)不確定條件下的數(shù)值仿真。同時統(tǒng)計平均控制量Fi和平均位置跟蹤誤差Ei以表征控制效果(下標1,2,3分別表示對比方法1、對比方法2和本文方法):
Fi=-∑Ns=1τ21,s+τ22,s+τ23,s3, i=1,2,3
Ei=∑Ns=1Δ2x,s+Δ2y,s+Δ2z,s3, i=1,2,3(60)
仿真結(jié)果如圖10~圖18和表1所示。由圖10可知3種方法均可實現(xiàn)AUV對三維軌跡的高精度跟蹤。對圖11~圖13展開分析可知,當(dāng)控制律中除去自適應(yīng)項時,抗擾動能力降低,使得對比方法2作用下的軌跡跟蹤誤差大于本文方法作用下的跟蹤誤差,而文獻[14]中跟蹤精度最低,這是由于漸進收斂方式的收斂速度慢,且抗擾性較預(yù)設(shè)時間收斂方式弱。表1中,本文方法作用下平均位置誤差320.35 m,對比方法2為320.81 m,而對比方法1則為493.321 m。觀察圖14和圖15知,航行全程兩種預(yù)設(shè)時間控制方法均實現(xiàn)良好的姿態(tài)角變化動態(tài),而文獻[14]方法在初期存在嚴重的姿態(tài)角振蕩現(xiàn)象,反映了漸進收斂方式下較差的暫態(tài)性能。相應(yīng)的,文獻[14]方法的控制量較大,而兩種預(yù)設(shè)時間收斂控制方法的控制量較小,其中由于本文方法使用自適應(yīng)律避免了對擾動的過估計,使得本文方法的控制量消耗最小,如圖16~圖18和表1所示。
4 結(jié) 論
針對欠驅(qū)動AUV在復(fù)雜擾動和參數(shù)不確定條件下的高性能三維軌跡跟蹤需求,本文結(jié)合動態(tài)過程函數(shù)和預(yù)設(shè)時間控制理論,提出動態(tài)性能及收斂時間預(yù)定義的抗擾軌跡跟蹤控制方法,可通過調(diào)節(jié)動態(tài)過程函數(shù)實現(xiàn)期望性能,并調(diào)節(jié)單個參數(shù)準確預(yù)設(shè)實際收斂時間,在滿足嚴格時間約束的同時,避免了現(xiàn)有預(yù)設(shè)時間控制方法預(yù)設(shè)保守性大及噪聲條件下實用性降低的問題。數(shù)值仿真結(jié)果表明,所提控制方法對參數(shù)不確定和外部擾動具有良好的魯棒性和抗擾性,且可實現(xiàn)良好的軌跡跟蹤動態(tài)性能。后續(xù)研究中,將針對本文未能解決的動態(tài)過程函數(shù)整定問題進一步展開研究,以期顯著提升控制方法的實用性。
參考文獻
[1] LI Z F, WANG M, MA G. Adaptive optimal trajectory tracking control of AUVs based on reinforcement learning[J]. ISA Transactions, 2023, 137: 122-132.
[2] SHEN C, SHI Y, BUCKHAM B. Path-following control of an AUV: a multiobjective model predictive control approach[J]. IEEE Trans.on Control Systems Technology, 2018, 27(3): 1334-1342.
[3] ZHANG G C, HUANG H, QIN H D, et al. A novel adaptive second order sliding mode path following control for a portable AUV[J]. Ocean Engineering, 2018, 151: 82-92.
[4] REZAZADEGAN F, SHOJAEI K, SHEIKHOLESLAM F, et al. A novel approach to 6-DOF adaptive trajectory tracking control of an AUV in the presence of parameter uncertainties[J]. Ocean Engineering, 2015, 107: 246-258.
[5] ALI N, TAWIAH I, ZHANG W. Finite-time extended state observer based nonsingular fast terminal sliding mode control of autonomous underwater vehicles[J]. Ocean Engineering, 2020, 218: 108179.
[6] ZHANG Z Y, LIN M W, LI D J. A double-loop control framework for AUV trajectory tracking under model parameters uncertainties and time-varying currents[J]. Ocean Engineering, 2022, 265: 112566.
[7] WU H M, KARKOUB M. Hierarchical backstepping control for trajectory-tracking of autonomous underwater vehicles subject to uncertainties[C]∥Proc.of the IEEE 14th International Conference on Control, Automation and Systems, 2014: 1191-1196.
[8] 周鑄, 李文魁, 呂志彪, 等. 擾動不確定的AUV改進反步控制[J]. 艦船電子工程, 2022, 42(12): 169-174.
ZHOU Z, LI W K, LYU Z B, et al. Improved backstepping control of uncertain AUVs under perturbations[J]. Ship Electronic Engineering, 2022, 42(12): 169-174.
[9] 李娟, 王佳奇, 丁福光. 基于反饋線性化的AUV三維軌跡跟蹤滑模控制[J]. 哈爾濱工程大學(xué)學(xué)報, 2022, 43(3): 348-355.
LI J, WANG J Q, DING F G. 3-D trajectory tracking sliding mode control of AUV based on feeedback linearization[J]. Journal of Harbin Engineering University, 2022, 43(3): 348-355.
[10] 李鑫濱, 王鵬, 駱曦, 等. 輸入受限下欠驅(qū)動AUV軌跡跟蹤滑模控制[J]. 水下無人系統(tǒng)學(xué)報, 2022, 30(1): 44-53.
LI X B, WANG P, LUO X, et al. Trajectory tracking sliding mode control of underactuated AUV with input constraints[J]. Journal of Underwater Unmanned Systems, 2022, 30(1): 44-53.
[11] LI J, DU J L, CHEN C L P. Command-filtered robust adaptive NN control with the prescribed performance for the 3-D trajectory tracking of underactuated AUVs[J]. IEEE Trans.on Neural Networks and Learning Systems, 2021, 33(11): 6545-6557.
[12] ZHANG J L, XIANG X B, ZHANG Q, et al. Neural network-based adaptive trajectory tracking control of underactuated AUVs with unknown asymmetrical actuator saturation and unknown dynamics[J]. Ocean Engineering, 2020, 218: 108193.
[13] 劉用, 楊曉飛, 夏金銘. 基于模糊算法的AUV避障與姿態(tài)控制[J]. 江蘇大學(xué)學(xué)報(自然科學(xué)版), 2021, 42(6): 655-660.
LIU Y, YANG X F, XIA J M. Obstacle-avoidance and attitude control of AUV based on fuzzy algorithm[J]. Journal of Jiangsu University (Natural Science Edition), 2021, 42(6): 655-660.
[14]" LIANG X, QU X R, WANG N, et al. Three-dimensional trajectory tracking of an underactuated AUV based on fuzzy dynamic surface control[J]. IET Intelligent Transport Systems, 2020, 14(5): 364-370.
[15] CHEN H X, TANG G Y, WANG S F, et al. Adaptive fixed-time backstepping control for three-dimensional trajectory tracking of underactuated autonomous underwater vehicles[J]. Ocean Engineering, 2023, 275: 114109.
[16] ZHENG J Q, SONG L, LIU L Y, et al. Fixed-time extended state observer-based trajectory tracking control for autonomous underwater vehicles[J]. Asian Journal of Control, 2022, 24(2): 686-701.
[17] SUN H B, ZONG G D, CUI J W, et al. Fixed-time sliding mode output feedback tracking control for autonomous underwater vehicle with prescribed performance constraint[J]. Ocean Engineering, 2022, 247: 110673.
[18] MOULAY E, LECHAPPE V, BERNUAU E, et al. Fixed-time sliding mode control with mismatched disturbances[J]. Automatica, 2022, 136: 110009.
[19] ZHENG J Q, SONG L, LIU L Y, et al. Fixed-time sliding mode tracking control for autonomous underwater vehicles[J]. Applied Ocean Research, 2021, 117: 102928.
[20] WANG H B, SU B, WANG Y L, et al. Fixed-time stabilization control for underactuated AUV with external disturbance[C]∥Proc.of the IEEE Chinese Control Conference, 2019: 4513-4518.
[21] AN S, WANG X Y, WANG L J, et al. Observer based fixed-time integral sliding mode tracking control for underactuated AUVs via an event-triggered mechanism[J]. Ocean Engineering, 2023, 284: 115158.
[22] LIU Y, LIU X P, JING Y W. Adaptive neural networks finite-time tracking control for non-strict feedback systems via prescribed performance[J]. Information Sciences, 2018, 468: 29-46.
[23] SUN Y C, ZHANG Y, QIN H D, et al. Predefined-time prescribed performance control for AUV with improved performance function and error transformation[J]. Ocean Engineering, 2023, 281: 114817.
[24] LI Y, HE J Y, ZHANG Q, et al. Predefined-time fault-tolerant trajectory tracking control for autonomous underwater vehicles considering actuator saturation[J]. Actuators, 2023, 12(4): 171-192.
[25] LI K W, LI Y M. Adaptive predefined-time optimal tracking control for underactuated autonomous underwater vehicles[J]. IEEE/CAA Journal of Automatica Sinica, 2023, 10(4): 1083-1085.
[26] ZHANG L, JU X Z, CUI N G. Ascent control of heavy-lift launch vehicle with guaranteed predefined performance[J]. Aerospace Science and Technology, 2021, 110: 106511.
[27] ZHOU H P, ZHENG Z W, GUAN Z Y, et al. Control barrier function based nonlinear controller for automatic carrier landing[C]∥Proc.of the IEEE 16th International Conference on Control, Automation, Robotics and Vision, 2020: 584-589.
[28] JU X Z, JIANG Y S, JING L, et al. Quantized predefined-time control for heavy-lift launch vehicles under actuator faults and rate gyro malfunctions[J]. ISA transactions, 2023, 138: 133-150.
[29] JU X Z, WEI C Z, XU H C, et al. Fractional-order sliding mode control with a predefined-time observer for VTVL reusable launch vehicles under actuator faults and saturation constraints[J]. ISA transactions, 2022, 129: 55-72.
[30] ZHANG L, LI D Y, JING L, et al. Appointed-time cooperative guidance law with line-of-sight angle constraint and time-to-go control[J]. IEEE Trans.on Aerospace and Electronic Systems, 2023, 59(3): 3142-3155.
[31] CHEN Z R, JU X Z, WANG Z W, et al. The prescribed time sliding mode control for attitude tracking of spacecraft[J]. Asian Journal of Control, 2022, 24(4): 1650-1662.
[32] HARDY G H, LITTLEWOOD J E, PLYA G. Inequalities[M]. Cambridge: Cambridge university press, 1952.
[33] YANG M, ZHANG Q, XU K, et al. Adaptive differentiator-based predefined-time control for nonlinear systems subject to pure-feedback form and unknown disturbance[J]. Complexity, 2021, 2021: 7029058.
[34] NI J K, SHI P. Global predefined time and accuracy adaptive neural network control for uncertain strict-feedback systems with output constraint and dead zone[J]. IEEE Trans.on Systems, Man, and Cybernetics-Systems, 2020, 51(12): 7903-7918.
[35] LIU B J, WANG W C, LI Y K, et al. Adaptive quantized predefined-time backstepping control for nonlinear strict-feedback systems[J]. IEEE Trans.on Circuits and Systems II: Express Briefs, 2022, 69(9): 3859-3863.
[36] PETTERSEN K Y, EGELAND O. Time-varying exponential stabilization of the position and attitude of an underactuated autonomous underwater vehicle[J]. IEEE Trans.on Automatic Control, 1999, 44(1): 112-115.
作者簡介
李曉斌(1988—),男,工程師,本科,主要研究方向為艦船控制、多艦船編隊控制。
徐 東(1982—),男,工程師,博士,主要研究方向為艦船控制、可靠性系統(tǒng)工程。
楊 雪(1988—),女,工程師,主要研究方向為海洋經(jīng)濟學(xué)。