含齿隙与饱和电动舵机的自适应神经网络输出反馈控制方法

doi:10.12382/bgxb.2025.0398

摘要/Abstract

摘要：

针对电动舵机控制器设计过程中的齿隙、输入饱和与状态信息不完全可测等问题，提出了一种基于自适应神经网络状态观测器的输出反馈控制方法。为刻画齿隙对系统动力学的影响，通过引入近似死区函数构建含齿隙的4阶伺服系统模型。针对状态不完全可测的问题，设计一种基于自适应神经网络的状态观测器，实现了存在模型不确定性条件下的系统状态重构。在反步法框架下，利用双曲正切Lyapunov函数结合状态观测器输出构建了输出反馈控制器。针对可能出现的输入饱和，引入辅助滤波系统以补偿输入饱和的影响。基于Lyapunov理论证明了闭环系统中误差信号的有界性。通过构建多组仿真实验，验证了所设计控制方法的有效性。研究结果表明，所设计的控制方法能够抑制齿隙非线性对系统性能的影响，补偿输入饱和约束，并在状态不完全可测条件下实现舵面的精确跟踪控制。

关键词: 电动舵机, 齿隙非线性, 自适应神经网络状态观测器, 输出反馈控制, 控制输入饱和

Abstract:

Aiming at the problems of backlash，input saturation and unmeasured state information in the design process of electromechanical actuator controller，an output feedback control method based on adaptive neural network state observer is proposed.In order to describe the influence of backlash on system dynamics，a fourth-order servo system model with backlash is constructed by introducing an approximate dead zone function.A state observer based on adaptive neural network is designed for the unmeasurable state to realize the reconstruction of system state under the condition of model uncertainty.Under the framework of backstepping method，an output feedback controller is constructed by using the hyperbolic tangent Lyapunov function and state observer output.For the possible input saturation，an auxiliary filtering system is introduced to compensate the influence of input saturation.The boundedness of error signals in the closed-loop system is proved based on Lyapunov theory.The effectiveness of the proposed control method is verified by conducting multiple sets of simulation experiments.The results demonstrate the superior performance of the proposed method in suppressing the effect of backlash nonlinearityt on the system performance，compensating for input saturation，and achieving the actuator accurate tracking control with unmeasured state information.

Key words: electromechanical actuator, backlash nonlinearity, adaptive neural network state observer, output feedback control, control input saturation

朱泽军, 王伟, 林时尧, 纪毅. 含齿隙与饱和电动舵机的自适应神经网络输出反馈控制方法[J]. 兵工学报, 2025, 46(S1): 250398-.

ZHU Zejun, WANG Wei, LIN Shiyao, JI Yi. Adaptive Neural Network Output Feedback Control Method for Electromechanical Actuator with Backlash and Saturation[J]. Acta Armamentarii, 2025, 46(S1): 250398-.

图/表 9

图1 舵机机电位置伺服系统结构

Fig.1 Structure of electromechanical position servo system for an actuator

图2 自适应神经网络输出反馈控制方案

Fig.2 Output feedback control scheme of adaptive neural network

表1 含齿隙电动舵机标称参数

Table 1 Nominal parameters of electromechanical actuator with backlash

参数	数值	参数	数值
J_m/(kg·m²)	0.05	K_e/(V·s·rad^-1)	0.11
J_d/(kg·m²)	0.16	K_t/(N·m·A^-1)	0.063
b_m/(N·m·s·rad^-1)	1.5	R/Ω	0.5
b_d/(N·m·s·rad^-1)	1.6	θ_dmax/(°)	12
i	5	h/(N·m·rad^-1)	0.12
α/rad	0.005	K_PWM	8
k/(N·m·rad^-1)	70

表2 神经网络以及自适应控制的参数

Table 2 Neural network and adaptive control parameters

参数	数值	参数	数值
π₂	5	γ₂	10
π₄	0.1	γ₄	10
π_g	0.5	γ_g	100
${\Lambda }_{{\beta }_{2}}$	10	${\gamma }_{{\beta }_{2}}$	1
${\Lambda }_{{\beta }_{4}}$	5	${\gamma }_{{\beta }_{4}}$	1
${\overline{\pi }}_{2}$	10	${\overline{\gamma }}_{2}$	1
κ	1	c₂，c₄	1

表3 PID、状态反馈和输出反馈控制器参数

Table 3 PID，state feedback and output feedback controller parameters

控制器	控制增益
PID	k_p		k_i		k_d
PID	-50		-100		1
状态反馈	K₁	K₂		K₃		K₄
状态反馈	15	30		2		5
输出反馈	k₁	k₂		k₃		k₄
输出反馈	15	30		2		5

图3 PID、状态反馈和输出反馈控制下对阶跃期望信号的跟踪情况

Fig.3 Tracking of expected step signal under PID，state feedback and output feedback control

图4 PID、状态反馈和输出反馈控制下对方波期望信号的跟踪情况

Fig.4 Tracking of expected square wave signal under PID，state feedback and output feedback control

图5 PID、状态反馈和输出反馈控制下对正弦期望信号的跟踪情况

Fig.5 Tracking of expected sine signal under PID，state feedback and output feedback control

图6 状态观测与自适应参数收敛特性

Fig.6 State observation and adaptive parameter convergence characteristics

参考文献 27

[1]	田福庆, 姜尚, 梁伟阁. 含齿隙弹载舵机的全局反步模糊自适应控制[J]. 自动化学报, 2019, 45(6):1177-1185.
	TIAN F Q, JIANG S, LIANG W G. Global backstepping fuzzy adaptive control for ammunition actuator with backlash[J]. Acta Automatica Sinica, 2019, 45(6):1177-1185. (in Chinese)
[2]	杨一帆, 张文静, 张金鹏, 等. 基于分数阶的电动伺服系统控制器设计[J]. 航空兵器, 2018(6):50-54.
	YANG Y F, ZHANG W J, ZHANG J P, et al. Controller design for electromechanical actuator servo system based on fractional-order[J]. Aero Weaponry, 2018(6):50-54. (in Chinese)
[3]	刘源翔, 崔岩, 李睿, 等. 制导炮弹可控滚转执行机构模型分析[J]. 弹道学报, 2017, 29(1):17-21.
	LIU Y X, CUI Y, LI R, et al. Modeling and analysis of controllable-rolling actuator system[J]. Journal of Ballistics, 2017, 29(1):17-21. (in Chinese)
[4]	殷希梅, 张航. 简谈中、小型制导弹药舵机的发展现状[J]. 兵工自动化, 2019, 38(4):71-74.
	YIN X M, ZHANG H. Discussion on development status of medium and small guided missile servo system[J]. Ordnance Industry Automation, 2019, 38(4):71-74. (in Chinese)
[5]	杜仁慧, 吴益飞, 陈威, 等. 考虑齿隙伺服系统的反步自适应模糊控制[J]. 控制理论与应用, 2013, 30(2):254-260.
	DU R H, WU Y F, CHEN W, et al. Adaptive backstepping fuzzy control for servo systems with backlash[J]. Control Theory & Applications, 2013, 30(2):254-260. (in Chinese)
[6]	吕帅帅, 林辉, 陈晓雷, 等. 弹载电动舵机幂次滑模反演控制[J]. 北京理工大学学报, 2016, 36(10):1037-1042.
	LV S S, LIN H, CHEN X L, et al. Backstepping power fast terminal siding mode control for missile borne electric rudder[J]. Transactions of Beijing Institute of Technology, 2016, 36(10):1037-1042. (in Chinese)
[7]	YU Z Y, TANG T. Limited amplitude-disturbance observer control for backlash-containing electromechanical system[J]. IEEE Transactions on Industrial Electronics, 2024, 71(6):5960-5971.
[8]	HE X Y, ZHAO Z J, SU J Y, et al. Adaptive inverse control of a vibrating coupled vessel-riser system with input backlash[J]. IEEE Transactions on Systems,Man,and Cybernetics:Systems, 2021, 51(8):4706-4715.
[9]	ZHANG P, CHEN Y, WAN Z, et al. Adaptive finite-time backstepping sliding mode control for PMSM system with backlash[J]. IEEE Journal of Emerging and Selected Topics in Power Electronics, 2022, 10(6):7549-7559.
[10]	安凯. 关节齿隙对空间机械臂末端定位精度的影响[J]. 兵工学报, 2014, 35(8):1301-1307. doi: 10.3969/j.issn.1000-1093.2014.08.025
	AN K. Effect of gear backlash on positioning precision of manipulator end-effectors[J]. Acta Armamentarii, 2014, 35(8):1301-1307. doi: 10.3969/j.issn.1000-1093.2014.08.025
[11]	TARBOURIECH S, QUEINNEC I, PRIEUR C. Stability analysis and stabilization of systems with input backlash[J]. IEEE Transactions on Automatic Control, 2014, 59(2):488-494.
[12]	SUN G F, ZHAO J B, CHEN Q. Observer-based compensation control of servo systems with backlash[J]. Asian journal of control, 2021, 23(1):499-512.
[13]	LAI G Y, LIU Z, ZHANG Y, et al. Adaptive inversion-based fuzzy compensation control of uncertain pure-feedback systems with asymmetric actuator backlash[J]. IEEE Transactions on Fuzzy Systems, 2017, 25(1):141-155.
[14]	BRAHMI B, SAAD M, OCHOA-LUNA C, et al. Adaptive tracking control of an exoskeleton robot with uncertain dynamics based on estimated time-delay control[J]. IEEE/ASME Transactions on Mechatronics, 2018, 23(2):575-585.
[15]	BAI C, YIN Z G, LI T, et al. Backstepping control collaborative shaft torque observer for limit cycle oscillation suppression of fully closed-loop gear transmission servo system[J]. IEEE Transactions on Power Electronics, 2024, 39(4):4513-4526.
[16]	ZHAO Z J, FENG K L, WANG X W, et al. Adaptive NN control for a flexible manipulator with input backlash and output constraint[J]. IEEE Transactions on Systems,Man,and Cybernetics:Systems, 2024, 54(12):7472-7481.
[17]	ZHAO Z J, HE W T, ZHANG F K, et al. Deterministic learning from adaptive neural network control for a 2-DOF helicopter system with unknown backlash and model uncertainty[J]. IEEE Transactions on Industrial Electronics, 2023, 70(9):9379-9389.
[18]	MEI Y F, LIU Y, WANG H. Adaptive neural network output-constraint control for a variable-length rotary arm with input backlash nonlinearity[J]. IEEE Transactions on Neural Networks and Learning Systems, 2023, 34(8):4741-4749.
[19]	茂隆, 孙秀霞, 徐光智, 等. 执行器非线性超低空空投航迹倾角自适应控制[J]. 电机与控制学报, 2017, 21(3):111-118.
	LV M L, SUN X X, XU G Z, et al. Adaptive controller for ultra-low altitude airdrop flight path angle with actuator nonlinearity[J]. Electric Machines and Control, 2017, 21(3):111-118. (in Chinese)
[20]	JING W J, LI M, CHEN Y, et al. Fixed-time control of multi-motor nonlinear systems via adaptive neural network dual sliding mode[J]. Information Sciences, 2025, 708:122061.
[21]	WANG X T, MEI X S, YANG J K, et al. Adaptive neural sliding mode control of an uncertain permanent magnet linear motor system with unknown input backlash in laser processing[J]. Information Sciences, 2024, 680:121087.
[22]	宋秋雨, 胡健, 姚建勇, 等. 面向输出约束基于神经网络观测器的发射平台输出反馈控制[J]. 兵工学报, 2023, 44(7):2184-2196. doi: 10.12382/bgxb.2022.0229
	SONG Q Y, HU J, YAO J Y, et al. Output feedback control for launch platform using neural network observer and output constraint[J]. Acta Armamentarii, 2023, 44(7):2184-2196. (in Chinese) doi: 10.12382/bgxb.2022.0229
[23]	YANG Y, HUANG J S, SU X J, et al. Event-triggered adaptive output feedback control of multivariable systems with nonsmooth actuator nonlinearities[J]. IEEE Transactions on Systems,Man,and Cybernetics:Systems, 2021, 51(9):5557-5566.
[24]	BAI Y C, HU J, YAO Y J. Adaptive neural network output feedback robust control of electromechanical servo system with backlash compensation and disturbance rejection[J]. Mechatronics, 2022, 84:102794.
[25]	TONG S C, LI Y M, LIU Y J. Observer-based adaptive neural networks control for large-scale interconnected systems with nonconstant control gains[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021, 32(4):1575-1585.
[26]	WANG C H, SHI Y, WANG Y X, et al. Event-triggered adaptive fuzzy output feedback tracking control for pneumatic servo system with input voltage saturation and position constraint[J]. IEEE Transactions on Industrial Informatics, 2023, 20(3):4360-4369.
[27]	REN Y, LIU Z J, ZHAO Z J, et al. Adaptive active anti-vibration control for a 3-D helicopter flexible slung-load system with input saturation and backlash[J]. IEEE Transactions on Aerospace and Electronic Systems, 2024, 60(1):320-333.