









摘要:針對具有未知參數(shù)的拉格朗日系統(tǒng),本文研究了安全跟蹤控制問題。首先采用backstepping方法,設(shè)計跟蹤控制器用于解決系統(tǒng)的軌跡跟蹤問題。其次構(gòu)造了自適應(yīng)控制障礙函數(shù),通過該函數(shù)得到的安全控制器確保系統(tǒng)狀態(tài)遠離安全集的邊界。然后結(jié)合所提出的跟蹤控制器以及自適應(yīng)控制障礙函數(shù)約束產(chǎn)生了二次規(guī)劃,基于二次規(guī)劃設(shè)計安全跟蹤控制器,使得系統(tǒng)安全的同時盡可能地跟蹤到參考信號。最后仿真結(jié)果驗證了所提出方法的有效性。
關(guān)鍵詞:拉格朗日系統(tǒng);控制障礙函數(shù);二次規(guī)劃;安全;跟蹤控制
中圖分類號:O231.2 " " " " " " " "文獻標(biāo)志碼:A
由拉格朗日系統(tǒng)描述的機械系統(tǒng),如車擺系統(tǒng)[1]、機械臂[2]、直升機[3]等,在實際的工業(yè)生產(chǎn)中有著廣泛的應(yīng)用。通常控制工程師們所需考慮的問題是如何設(shè)計控制器使得系統(tǒng)具有良好的性能。但隨著現(xiàn)代控制的發(fā)展,大多數(shù)系統(tǒng)需要在現(xiàn)實環(huán)境中工作,因此系統(tǒng)的安全性受到廣泛關(guān)注。一般來說,安全約束可能與系統(tǒng)固有的特性或系統(tǒng)的交互環(huán)境有關(guān)。
控制障礙函數(shù)是確保系統(tǒng)安全的重要工具,其用于設(shè)計安全控制器,從而使得系統(tǒng)狀態(tài)滿足嚴(yán)格的安全約束。文獻[4]提出的倒數(shù)控制障礙函數(shù)和歸零控制障礙函數(shù)確保了集合的前向不變性。文獻[5]給出了自適應(yīng)控制障礙函數(shù)使得具有未知參數(shù)的系統(tǒng)狀態(tài)遠離安全集的邊界。文獻[6]通過例子說明了如何構(gòu)造控制障礙函數(shù)使得系統(tǒng)狀態(tài)保持在給定的集合內(nèi)。文獻[7-10]提出了以控制障礙函數(shù)條件所構(gòu)成的二次規(guī)劃,基于二次規(guī)劃設(shè)計的控制器確保了系統(tǒng)安全。文獻[11]通過二次規(guī)劃給出了控制器的顯式解,這使得不必在線求解二次規(guī)劃并便于分析。但是上述文獻只考慮了系統(tǒng)安全,并未考慮系統(tǒng)性能。
目前自適應(yīng)backstepping[12-15]和滑模控制[16-19]等方法已廣泛應(yīng)用于實現(xiàn)系統(tǒng)的跟蹤控制性能,然而這些結(jié)果只關(guān)注了系統(tǒng)的控制性能,而忽略了安全性問題。由于違反安全約束可能會給系統(tǒng)帶來不可逆的損壞,所以在確保系統(tǒng)滿足安全約束的同時考慮系統(tǒng)控制性能是有必要的。文獻[20]提出了一種基于安全自適應(yīng)強化學(xué)習(xí)的自主避障控制方法,使得系統(tǒng)實現(xiàn)避障問題的同時能夠跟蹤到參考信號。文獻[21]設(shè)計了具有避障功能的滑模控制器使得系統(tǒng)在避障的前提下跟蹤誤差能夠收斂到零的一個小鄰域內(nèi)。文獻[22]提出了自適應(yīng)滑模控制器,使得系統(tǒng)狀態(tài)一直保持在安全集內(nèi)并且跟蹤誤差收斂到零。然而基于backstepping控制方法設(shè)計安全跟蹤控制器使得系統(tǒng)安全并且跟蹤到參考信號的相關(guān)文獻較少。
基于以上討論,本文研究了具有未知參數(shù)的拉格朗日系統(tǒng)的安全跟蹤控制問題。主要工作包括以下幾個方面:
(1)針對具有未知參數(shù)的拉格朗日系統(tǒng),設(shè)計自適應(yīng)backstepping標(biāo)稱跟蹤控制器使得系統(tǒng)能夠跟蹤到參考信號,然后通過所提出的自適應(yīng)控制障礙函數(shù),基于backstepping方法設(shè)計安全控制器使得系統(tǒng)狀態(tài)保持在安全集內(nèi)。
(2)結(jié)合上述所提出的標(biāo)稱跟蹤控制器與自適應(yīng)控制障礙函數(shù)的約束生成了二次規(guī)劃,基于二次規(guī)劃給出了安全跟蹤控制器的顯式解,使得系統(tǒng)在保證安全的前提下盡可能地跟蹤到參考信號。
仿真結(jié)果如圖2-4所示。圖2只考慮安全沒有涉及跟蹤的軌跡,表明了自適應(yīng)障礙函數(shù)的值為正,在只考慮安全時系統(tǒng)狀態(tài) 一直保持在給定的范圍內(nèi)。圖3表明了 滿足安全約束并且跟蹤到給定的參考信號 。圖4表明了安全跟蹤時的跟蹤誤差軌跡收斂到零的小鄰域內(nèi),并且系統(tǒng)狀態(tài) 滿足安全約束。圖2-4表明了本文所提出的控制策略的有效性。
5 結(jié) 論
本文研究了具有未知參數(shù)的拉格朗日系統(tǒng)的安全跟蹤控制問題。基于backstepping設(shè)計標(biāo)稱跟蹤控制器使得系統(tǒng)能夠跟蹤給定的參考信號,提出自適應(yīng)控制障礙函數(shù)的約束來保證系統(tǒng)狀態(tài)在給定的安全集內(nèi)。結(jié)合上述標(biāo)稱跟蹤控制器及自適應(yīng)控制障礙函數(shù)約束產(chǎn)生了二次規(guī)劃,基于二次規(guī)劃設(shè)計的安全跟蹤控制器使得系統(tǒng)安全并且盡可能地跟蹤參考信號。仿真結(jié)果驗證了所提出方案的有效性。
目前還存在一些需要考慮的問題,如對于存在隨機擾動影響的高階系統(tǒng)如何設(shè)計控制器使得系統(tǒng)在安全的前提下盡可能地跟蹤到給定的參考信號。
參考文獻
[1] CHATTERJEE D, PATRA A, JOGLEKAR H K. Swing-up and stabilization of a cart-pendulum system under restricted cart track length[J]. Systems amp; Control Letters, 2002, 47(4): 355-364.
[2] MAHYUDDIN M N, KHAN S G, HERRMANN G. A novel robust adaptive control algorithm with finite-time online parameter estimation of a humanoid robot arm[J]. Robotics and Autonomous Systems, 2014, 62(3): 294-305.
[3] RAFFO G V, ORTEGA M G, RUBIO F R. An integral predictive/nonlinear H∞ control structure for a quadrotor helicopter[J]. Automatica, 2010, 46(1): 29-39.
[4] AMES A D, XU X R, GRIZZLE J W, et al. Control barrier function based quadratic programs for safety critical systems[J]. IEEE Transactions on Automatic Control, 2017, 62(8): 3861-3876.
[5] TAYLOR A J, AMES A D. Adaptive safety with control barrier functions[C]//2020 American Control Conference (ACC). "Denver: IEEE, 2020: 1399-1405.
[6] FERRAGUTI F, LANDI C T, SINGLETARY A, et al. Safety and efficiency in robotics: the control barrier functions approach[J]. IEEE Robotics amp; Automation Magazine, 2022, 29(3): 139-151.
[7] 陳杰, 呂梓亮, 黃鑫源,等. 非線性系統(tǒng)的安全分析與控制: 障礙函數(shù)方法[J]. 自動化學(xué)報, 2023, 49(3):567-579.
[8] XIONG X B, AMES A. SLIP walking over rough terrain via H-LIP stepping and backstepping-barrier function inspired quadratic program[J]. IEEE Robotics and Automation Letters, 2021, 6(2): 2122-2129.
[9] XU X R, TABUADA P, GRIZZLE J W, et al. Robustness of control barrier functions for safety critical control[J]. IFAC-PapersOnLine, 2015, 48(27): 54-61.
[10] FARZAN S, AZIMI V, HU A P, et al. Adaptive control of wire-borne underactuated brachiating robots using control Lyapunov and barrier functions[J]. IEEE Transactions on Control Systems Technology, 2022, 30(6): 2598-2614.
[11] JANKOVIC M. Robust control barrier functions for constrained stabilization of nonlinear systems[J]. Automatica, 2018, 96: 359-367.
[12] HE W, DONG Y T, SUN C Y. Adaptive neural impedace control of a robotic manipulator with input saturation[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2016, 46(3): 334-344.
[13] HE W, YAN Z C, SUN C Y, et al. Adaptive neural network control of a flapping wing micro aerial vehicle with disturbance observer[J]. IEEE Transactions on Cybernetics, 2017, 47(10): 3452-3465.
[14] ROY S, ROY S B, KAR I N. Adaptive–robust control of Euler–Lagrange systems with linearly parametrizable uncertainty bound[J]. IEEE Transactions on Control Systems Technology, 2018, 26(5): 1842-1850.
[15] ROY S, KAR I N, LEE J, et al. Adaptive-robust time-delay control for a class of uncertain Euler-Lagrange systems[J]. IEEE Transactions on Industrial Electronics, 2017, 64(9): 7109-7119.
[16] BEN ROMDHANE NEILA M, TARAK D. Adaptive terminal sliding mode control for rigid robotic manipulators[J]. International Journal of Automation and Computing, 2011, 8: 215-220.
[17] ANUAR H A, PLESTAN F, CHRIETTE A, et al. Sliding mode control with adaptive gain of quadrotor with rigid manipulator[C]//2021 20th International Conference on Advanced Robotics (ICAR). Ljubljana, Slovenia: IEEE, 2022:547-554.
[18] EDWARDS C, SHTESSEL Y B. Adaptive continuous higher order sliding mode control[J]. Automatica, 2016, 65: 183-190.
[19] GAO D X, SUN Z Q, WANG W. Adaptive fuzzy sliding mode control for robotic manipulators[C]//2010 8th World Congress on Intelligent Control and Automation. Jinan: IEEE, 2010: 4811-4816.
[20] 王珂, 穆朝絮, 蔡光斌,等. 基于安全自適應(yīng)強化學(xué)習(xí)的自主避障控制方法[J]. 中國科學(xué):信息科學(xué), 2022, 52(9):1672-1686.
[21] LI X H, SONG S M, GUO Y. Robust finite-time tracking control for Euler–Lagrange systems with obstacle avoidance[J]. Nonlinear Dynamics, 2018, 93(2): 443-451.
[22] ZHANG Z J, WANG Y C, WOLLHERR D. Safe tracking control of Euler-Lagrangian systems based on a novel adaptive super-twisting algorithm [J]. IFAC-PapersOnLine, 2020, 53(2): 9974-9979.
[23] KOLATHAYA S, AMES A D. Input-to-state safety with control barrier functions[J]. IEEE Control Systems Letters, 2019, 3(1): 108-113.
Safe Tracking Control of Lagrangian Systems with Unknown Parameters
LI Jiayin , ZHANG Hui
(School of Mathematics and Informational Sciences, Yantai University, Yantai 264005, China)
Abstract: This paper focuses on safe tracking control problem of Lagrangian systems with unknown parameters. Firstly, the tracking controller is designed for trajectory tracking by the backstepping method. Then an adaptive control barrier function is constructed to design the safety controller, such that the system state is far from the boundary of the safety set. According to the combination of tracking controller and adaptive control barrier function constraints, the quadratic program is generated. Based on the quadratic program, the safety tracking controller is designed to guarantee the system track the reference signal as closely as possible under the premise of safety. The simulation results are given to illustrate the effectiveness of the proposed method.
Key words: Lagrangian system; control barrier function; quadratic program; safety; tracking control