摘要:針對狀態(tài)不可測和存在外部未知擾動的非線性多智能體系統(tǒng)的一致跟蹤問題,提出一種基于神經(jīng)網(wǎng)絡的分布式自適應脈沖控制方案。首先,構建復合擾動觀測器,解決系統(tǒng)狀態(tài)不可測與外部未知擾動耦合作用下的系統(tǒng)狀態(tài)感知問題;然后,通過自適應脈沖更新律,實現(xiàn)神經(jīng)網(wǎng)絡權值參數(shù)的快速估計,提升系統(tǒng)的瞬態(tài)性能;接著,結合脈沖動態(tài)系統(tǒng)的Lyapunov穩(wěn)定性理論,證明了閉環(huán)系統(tǒng)的一致最終有界性;最后,通過多單臂機械手系統(tǒng)的仿真實驗,驗證了該方案的有效性及優(yōu)越性。
關鍵詞:非線性多智能體;徑向基函數(shù)神經(jīng)網(wǎng)絡;自適應控制;脈沖控制;觀測器
中圖分類號:TP13; TP183;"O231.2""""文獻標志碼:A """"""文章編號:1674-2605(2025)01-0003-10
DOI:10.3969/j.issn.1674-2605.2025.01.003""""""""nbsp;""""""""""""開放獲取
Adaptive Pulse Control of Nonlinear Multi-agent Systems """""""""""""""Based on Neural Networks
LUO Zhenfa
(Guangdong University of Technology, Guangzhou 510006,"China)
Abstract:"A distributed adaptive pulse control scheme based on neural networks is proposed for the consistent tracking problem of nonlinear multi-agent systems with unmeasurable states and external unknown disturbances. Firstly, construct a composite disturbance observer to solve the problem of system state awareness under the coupling of unmeasurable system states and external unknown disturbances. Then, by using an adaptive pulse update law, the neural network weight parameters can be quickly estimated to improve the transient performance of the system. Furthermore, based on the Lyapunov stability theory of pulse dynamic systems, it is proved that all signals in the closed-loop system are uniformly ultimately bounded. Finally, the effectiveness and superiority of the proposed scheme were verified through simulation experiments of a multi arm robotic arm system.
Keywords:"nonlinear multi-agent system; radial basis function neural network; adaptive control; pulse control; observer
0 引言
多智能體系統(tǒng)通過多個子系統(tǒng)之間的協(xié)同合作來完成各類復雜任務,廣泛應用于機器人、航天器和無人機等領域[1-3]。一致跟蹤控制作為多智能體系統(tǒng)協(xié)同合作的基本問題之一,吸引了大批學者開展研究,并取得了一定成果[4-6]。但這些研究大多集中于線性多智能體系統(tǒng)。對于非線性多智能體系統(tǒng),特別是不確定非線性多智能體系統(tǒng),其一致跟蹤控制沒有得到充分研究。
隨著人工智能技術的快速發(fā)展,神經(jīng)網(wǎng)絡因具有良好的非線性逼近能力,被廣泛應用于不確定非線性系統(tǒng)的自適應控制設計中。文獻[7]針對高階非線性多智能體系統(tǒng),提出一種基于觀測器的自適應神經(jīng)網(wǎng)絡一致跟蹤控制策略,解決了系統(tǒng)狀態(tài)不可測的問題。文獻[8]討論了不確定非線性系統(tǒng)的自適應神經(jīng)網(wǎng)絡輸出反饋控制問題,通過其設計的干擾觀測器,避免了未知擾動的影響。文獻[9]提出一種基于最小學習參數(shù)的分布式多智能體系統(tǒng)的自適應神經(jīng)網(wǎng)絡一致跟蹤控制協(xié)議,有效減少了在線學習的參數(shù)量。在自適應神經(jīng)網(wǎng)絡控制設計中,神經(jīng)網(wǎng)絡權值參數(shù)的估計十分重要,快速的自適應權值參數(shù)估計,可以改善系統(tǒng)的瞬態(tài)性能,獲得更好的控制效果。為此,文獻[10-11]設計了一種預估器來代替?zhèn)鹘y(tǒng)的動態(tài)面誤差,由于預估誤差具有額外的可調(diào)參數(shù),加快了神經(jīng)網(wǎng)絡權值參數(shù)的估計速率,但額外的預估器使系統(tǒng)控制結構更加復雜,并增加了計算負擔。因此,為了獲得更好的瞬態(tài)性能,仍需進一步研究自適應神經(jīng)網(wǎng)絡控制。
脈沖控制可以提高系統(tǒng)的控制性能,具有控制動作快、結構簡單、魯棒性強等特點,在實際系統(tǒng)工程中得到廣泛的研究與應用。文獻[12]設計了脈沖反饋控制律,對給定的參考信號具有較好的跟蹤效果。文獻[13]引入脈沖觀測器,通過合理利用原始輸出來改善跟蹤性能。文獻[14]設計了自適應脈沖反饋控制方案,有效提高了系統(tǒng)的同步性能。將脈沖控制與自適應神經(jīng)網(wǎng)絡控制相結合,可獲得更好的系統(tǒng)瞬態(tài)性能,這對控制理論的研究和應用具有重要意義。
本文針對狀態(tài)不可測和存在外部未知擾動的非線性多智能體系統(tǒng),設計一種基于神經(jīng)網(wǎng)絡的分布式自適應脈沖控制方案,以實現(xiàn)多智能體系統(tǒng)的一致跟蹤控制。首先,構建復合擾動觀測器,同時考慮了外部擾動和神經(jīng)網(wǎng)絡逼近誤差,提高了系統(tǒng)的控制性能;然后,提出一種自適應脈沖更新律,實現(xiàn)神經(jīng)網(wǎng)絡權值參數(shù)的快速估計;接著,基于反步遞推方法,設計自適應脈沖控制器;最后,建立一個脈沖動態(tài)系統(tǒng),利用擴展的Lyapunov穩(wěn)定性理論,證明了閉環(huán)系統(tǒng)的一致最終有界性。
1 相關內(nèi)容
1.1 圖論知識
1.2 問題描述
1.3 RBF神經(jīng)網(wǎng)絡
1.4 脈沖動態(tài)系統(tǒng)
2 方案設計
首先,設計復合擾動觀測器,用于處理系統(tǒng)不可測狀態(tài)與外部未知擾動;然后,基于反步遞推方法設計自適應脈沖控制器?;谏窠?jīng)網(wǎng)絡的分布式自適應脈沖控制方案設計示意圖如圖1所示。
2.1 復合擾動觀測器
本文設計的自適應脈沖更新律,可在不產(chǎn)生高頻振蕩信號的前提下,快速自適應估計RBF神經(jīng)網(wǎng)絡的權值參數(shù),從而提高多智能體系統(tǒng)的狀態(tài)觀測速率,改善系統(tǒng)瞬態(tài)性能。
2.2 自適應脈沖控制器
3 穩(wěn)定性分析
首先,基于神經(jīng)網(wǎng)絡的分布式自適應脈沖控制方案建立一個脈沖動態(tài)系統(tǒng),并給出主要的穩(wěn)定性結果;然后,分別分析脈沖間隔動態(tài)和脈沖動態(tài);最后,給出閉環(huán)系統(tǒng)的穩(wěn)定性證明。
4 仿真驗證
通過多單臂機械手系統(tǒng)的仿真實驗,驗證本文提出的基于神經(jīng)網(wǎng)絡的分布式自適應脈沖控制方案的有效性,并與文獻[25]方案進行對比實驗,驗證本文方案的優(yōu)越性。
在MATLAB平臺上,利用本文提出的基于神經(jīng)網(wǎng)絡的分布式自適應脈沖控制方案對上述多單臂機械手系統(tǒng)進行仿真,結果如圖3~9所示。其中,圖8、9為本文方案與文獻[25]方案的仿真效果對比圖。
由圖8、9可以看出,在相同的設計參數(shù)下,本文方案相較于文獻[25]方案具有更快速的自適應能力,可快速估計神經(jīng)網(wǎng)絡權值參數(shù),從而提高狀態(tài)觀測速率,進一步改善系統(tǒng)的瞬態(tài)性能。
5 結論
本文針對狀態(tài)不可測和存在外部未知擾動的非線性多智能體系統(tǒng),提出一種基于神經(jīng)網(wǎng)絡的分布式自適應脈沖控制方案,以解決系統(tǒng)的一致跟蹤控制問題。本文提出的復合擾動觀測器同時考慮了外部擾動和神經(jīng)網(wǎng)絡逼近誤差,提高了系統(tǒng)的控制性能;自適應脈沖更新律快速實現(xiàn)神經(jīng)網(wǎng)絡權值參數(shù)的收斂,改善了閉環(huán)系統(tǒng)的瞬態(tài)性能;在脈沖動態(tài)系統(tǒng)中,通過擴展的Lyapunov穩(wěn)定性理論證明了閉環(huán)系統(tǒng)的一致最終有界性。通過多單臂機械手系統(tǒng)的仿真對比實驗,驗證了本文方案的有效性和優(yōu)越性。但本文方案對于其他類型的系統(tǒng),如隨機非線性系統(tǒng)、分數(shù)階系統(tǒng)、切換非線性系統(tǒng),是否具有普適性還有待進一步驗證。同時,本文只考慮了系統(tǒng)狀態(tài)不可測和存在外部擾動的情況,對于存在約束、控制增益未知的系統(tǒng)未進行研究,后續(xù)開展擴展研究十分必要。
?The author(s) 2024. This is an open access article under the CC BY-NC-ND 4.0 License (https://creativecommons.org/licenses/ by-nc-nd/4.0/)
參考文獻
[1] DU H, WEN G, CHENG Y, et al. Distributed finite-time cooperative control of multiple high-order nonholonomic mo-bile robots[J]. IEEE Transactions on Neural Networks and Learning Systems, 2016,28(12):2998-3006.
[2] DU H, LI S. Attitude synchronization for flexible spacecraft with communication delays[J]. IEEE Transactions on Automatic Control, 2016,61(11):3625-3630.
[3] DONG X, HUA Y, ZHOU Y, et al. Theory and experiment on formation-containment control of multiple multirotor un-manned aerial vehicle systems[J]. IEEE Transactions on Auto-mation Science and Engineering, 2018,16(1):229-240.
[4] WANG X, WANG H, LI C, et al. Consensus seeking in multiagent systems with Markovian switching topology under aperiodic sampled data[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2018,50(12):5189-5200.
[5] WANG L, XIAO F. Finite-time consensus problems for networks of dynamic agents[J]. IEEE Transactions on Auto-matic Control, 2010,55(4):950-955.
[6] YU W, ZHENG W X, CHEN G, et al. Second-order consensus in multi-agent dynamical systems with sampled position data[J]. Automatica, 2011,47(7):1496-1503.
[7] CHEN C L P, WEN G X, LIU Y J, et al. Observer-based adaptive backstepping consensus tracking control for high-order nonlinear semi-strict-feedback multiagent systems[J]. IEEE Transactions on Cybernetics, 2015,46(7):1591-1601.
[8] LIAN Y, XIA J, PARK J H, et al. Disturbance observer-based adaptive neural network output feedback control for uncertain nonlinear systems[J]. IEEE Transactions on Neural Networks and Learning Systems, 2022,34(10):7260-7270.
[9] SHANG Y, CHEN B, LIN C. Consensus tracking control for distributed nonlinear multiagent systems via adaptive neural backstepping approach[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2018,50(7):2436-2444.
[10] PENG Z, WANG D, WANG J. Predictor-based neural dynamic surface control for uncertain nonlinear systems in strict-feedback form[J]. IEEE Transactions on Neural Net-works and Learning Systems, 2016,28(9):2156-2167.
[11] YANG Y, LIU Q, YUE D, et al. Predictor-based neural dynamic surface control for bipartite tracking of a class of nonlinear multiagent systems[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021,33(4):1791-1802.
[12] TUMA T, PANTAZI A, LYGEROS J, et al. Nanopositioning with impulsive state multiplication: A hybrid control approach [J]. IEEE Transactions on Control Systems Technology, 2012, 21(4):1352-1364.
[13] YU H, CHEN T. Event-triggered tracking control with filtered outputs and impulsive observers[J]. IEEE Transactions on Cybernetics, 2020,52(6):4981-4992.
[14] ZHU D, WANG R, LIU C, et al. Synchronization of chaotic-oscillation permanent magnet synchronous generators networks via adaptive impulsive control[J]. IEEE Transactions on Circuits"and Systems II: Express Briefs, 2019,67(10):2194-2198.
[15] ZHOU Q, DU P, LI H, et al. Adaptive fixed-time control of error-constrained pure-feedback interconnected nonlinear sys-tems[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2020,51(10):6369-6380.
[16] SUI S, TONG S. Finite-time fuzzy adaptive PPC for nonstrict-feedback nonlinear MIMO systems[J]. IEEE Transactions on Cybernetics, 2022,53(2):732-742.
[17] CHEN D, LIU X, YU W, et al. Neural-network based adaptive self-triggered consensus of nonlinear multi-agent systems with sensor saturation[J]. IEEE Transactions on Network Science and Engineering, 2021,8(2):1531-1541.
[18] LU S, CHEN M, LIU Y, et al. Adaptive NN tracking control for uncertain MIMO nonlinear system with time-varying state constraints and disturbances[J]. IEEE Transactions on Neural Networks and Learning Systems, 2022,34(10):7309-7323.
[19] ZHANG Y, SUN J, LIANG H, et al. Event-triggered adaptive tracking control for multiagent systems with unknown distur-bances[J]. IEEE Transactions on Cybernetics, 2018,50(3):890-901.
[20] WANG X, WANG H, HUANG T, et al. Neural-network-based adaptive tracking control for nonlinear multiagent systems: The observer case[J]. IEEE Transactions on Cybernetics, 2021,"53(1):138-150.
[21] SUN Y, CHEN B, LIN C, et al. Adaptive neural control for a class of stochastic nonlinear systems by backstepping ap-proach[J]. Information Sciences, 2016,369:748-764.
[22] HADDAD W M, CHELLABOINA V S, NERSESOV S G. Impulsive and hybrid dynamical systems: stability, dissipa-tivity, and control[M]. Princeton University Press, 2006.
[23] KHALIL HK. Nonlinear systems third edition[M]. Upper Saddle River, NJ: Prentice Hall,"2002.
[24] CHEN B, LIU X, LIU K, et al. Direct adaptive fuzzy control of nonlinear strict-feedback systems[J]. Automatica, 2009,45(6): 1530-1535.
[25] CHEN M, GE S S. Adaptive neural output feedback control of uncertain nonlinear systems with unknown hysteresis using disturbance observer[J]. IEEE Transactions on Industrial Electronics, 2015,62(12):7706-7716.
作者簡介:
羅振發(fā),男,1999年生,碩士研究生,主要研究方向:自適應控制、神經(jīng)網(wǎng)絡。E-mail:"m18379124404@163.com