变学习强度扑翼飞行器自学习控制

张承玺; 卢瑞秋; 李权; 吴荩; 许德智

doi:10.16383/j.aas.c250664

变学习强度扑翼飞行器自学习控制

doi: 10.16383/j.aas.c250664 cstr: 32138.14.j.aas.c250664

张承玺^1,,
卢瑞秋^1,,
李权^2,,
吴荩^3,,
许德智^4,

1.
江南大学自动化与智能科学学院(物联网学院)无锡 214122
2.
江苏第二师范学院人工智能研究中心南京 210013
3.
北京科技大学智能科学与技术学院北京 100083
4.
东南大学电气工程学院, 电动载运技术教育部工程中心南京 210096

基金项目: 国家自然科学基金(62573211) 资助

详细信息

作者简介:
张承玺：江南大学自动化与智能科学学院(物联网学院)副教授. 2019年于上海交通大学大学获得博士学位. 主要研究方向为航天器集群协同控制与管理、机器人智能感知与自主决策. E-mail: cxzhang@jiangnan.edu.cn

卢瑞秋：江南大学自动化与智能科学学院(物联网学院)硕士研究生. 主要研究方向为航天器姿态控制系统故障诊断与容错控制. E-mail: luruiqiu@163.com

李权：江苏第二师范学院人工智能研究中心讲师. 2025年于江南大学获得博士学位. 主要研究方向为工业过程建模、智能优化控制及机器人感知与控制. 本文通信作者. E-mail: liquan95@jssnu.edu.cn

吴荩：北京科技大学智能科学与技术学院教授. 2024年于香港科技大学获得博士学位. 主要研究方向为机器人学、位姿估计及组合导航系统. E-mail: wujin@ustb.edu.cn

许德智：东南大学电气工程学院教授. 2013年于南京航空航天大学获得博士学位. 主要研究方向为数据驱动控制、故障诊断与容错控制、多智能体系统、信息物理系统、可再生能源技术、电机控制以及智能电网. E-mail: xudezhi@seu.edu.cn

计量
- 文章访问数: 10
- HTML全文浏览量: 4
- 被引次数: 0
出版历程
- 收稿日期: 2025-11-24
- 录用日期: 2026-03-26
- 网络出版日期: 2026-04-30

Self-Learning Control for Flapping-Wing Air Vehicles with Variable Learning Intensity

ZHANG Cheng-Xi^1
,,
LU Rui-Qiu^1
,,
LI Quan^2
,,
WU Jin^3
,,
XU De-Zhi^4
,

1.
School of Internet of Things Engineering, Jiangnan University, Wuxi 214122
2.
Research Center for Artificial Intelligence, Jiangsu Second Normal University, Nanjing 210013
3.
School of Intelligent Science and Technology, University of Science and Technology Beijing, Beijing 100083
4.
School of Electrical Engineering, Engineering Research Center of Electrical Transport Technology, Ministry of Education, Southeast University, Nanjing, 210096

Funds: Supported by National Natural Science Foundation of China (62573211)

More Information

Author Bio:
ZHANG Cheng-Xi　Associate Professor at the School of Internet of Things Engineering, Jiangnan University. He received the Ph. D. degree from Shanghai Jiao Tong University in 2019. His research covers spacecraft swarm cooperative control and management, and robot intelligent perception and autonomous decision-making

LU Rui-Qiu　Master student at the School of Internet of Things Engineering, Jiangnan University. His research interest covers fault diagnosis and fault-tolerant control of spacecraft attitude control systems

WU Jin　Professor at the School of Intelligence Science and Technology, University of Science and Technology Beijing. He received the Ph. D. degree from the Hong Kong University of Science and Technology in 2024. His research interest covers robotics, pose estimation, and integrated navigation systems

XU De-Zhi　Professor at the School of Electrical Engineering, Southeast University. He received the Ph. D. degree from the Nanjing University of Aeronautics and Astronautics in 2013. His research interest covers data-driven control, fault diagnosis and fault-tolerant control, multi-agent systems, cyber-physical systems, technologies of renewable energy, motor control, and smart grid

摘要

摘要: 针对扑翼飞行器存在非线性动态、模型不确定性及嵌入式平台算力受限等问题, 本文提出一种可自定义变学习强度自学习控制方法. 该方法通过学习历史控制信息, 仅基于一个代数方程, 避免复杂控制器设计并有效提升轨迹跟踪精度与系统鲁棒性. 针对扑翼系统, 使用自定义函数对学习强度进行调节, 提高系统动态响应速度与稳态性能. 仿真结果表明, 所提方法在保持低计算复杂度同时, 具有优越控制性能.
- 扑翼飞行器 /
- 自学习控制 /
- 变学习强度
Abstract: To address the issues of nonlinear dynamics, model uncertainties, and limited computational resources on embedded platforms in flapping-wing air vehicles (FWAVs), this paper proposes a customizable variable learning intensity (VLI) self-learning control method. The proposed method learns from historical control information and is based on a single algebraic equation, avoiding complex controller designs while effectively improving trajectory tracking accuracy and system robustness. For the flapping-wing system, a custom function is used to adjust the learning intensity, enhancing the system's dynamic response speed and steady-state performance. Simulation results demonstrate that the proposed method maintains low computational complexity while achieving superior control performance.
- Flapping-wing aerial vehicle /
- self-learning control /
- variable learning intensity

HTML全文

图 1 扑翼飞行器在三维空间中飞行轨迹对比

Fig. 1 Comparison of the Flight Trajectories of Flapping-Wing Air Vehicles in 3D Space

下载: 全尺寸图片幻灯片

图 2 扑翼飞行器水平控制力

Fig. 2 The Horizontal Control Forces of Flapping-Wing Air Vehicles

下载: 全尺寸图片幻灯片

图 3 扑翼飞行器飞行轨迹跟踪误差

Fig. 3 Trajectory Tracking Error of Flapping-Wing Air Vehicles

下载: 全尺寸图片幻灯片

图 4 变学习强度曲线

Fig. 4 Variable Learning Intensity Curve

下载: 全尺寸图片幻灯片

表 1 仿真参数设置

Table 1 Simulation Parameter Settings

参数	数值	参数	数值
$m$	0.5 kg	$k_{p,\; \text{pos}}$	1.2
$g$	9.81 m/s	$k_{d,\; \text{pos}}$	1.3
$c_D $	0.10 s$^{-1}$	$k_{v}$	3.2
$c$	26	$R$	6.0 m
$n_p$	17	$\omega$	0.20 rad/s
$L_p$	0.04 H	$z_{\text{ref}}$	5.0 m
$R_p$	0.08 $\Omega$	$k_{p,\; z}$	4.0
$\beta$	$0.001$ N	$k_{d,\; z}$	3.0
$k_e$	0.06 V$\cdot$s/rad	$k_{i,\; \Theta}$	40
$k_L$	2.2 N/rad	$k_{p,\; \Theta}$	2.5
$k_T$	0.65 N/rad	$\alpha_{ff}$	0.592

下载: 导出CSV

表 2 控制算法计算性能指标对比

Table 2 Computational Performance Comparison of Control Algorithms

性能指标	文中SLC	增广状态观测器	自适应控制
存储空间(Bytes)	40	1152	1120
浮点运算(FLOPs)	9	162	240
单次计算耗时($\mu$s)	0.2512	1.3917	11.3533

下载: 导出CSV

参考文献(25)

[1]	贺威, 丁施强, 孙长银. 扑翼飞行器的建模与控制研究进展. 自动化学报, 2017, 43(05): 685−696 doi: 10.16383/j.aas.2017.c160581 He Wei, Ding Shi-Qiang, Sun Chang-Yin. Research Progress on Modeling and Control of Flapping-wing Air Vehicles. Acta Automatica Sinica, 2017, 43(05): 685−696 doi: 10.16383/j.aas.2017.c160581
[2]	Zhou C, Zhang Y, Wu J H. Effect of flexibility on unsteady aerodynamic forces of a purely plunging airfoil. Chinese Journal of Aeronautics, 2020, 33(1): 88−101 doi: 10.1016/j.cja.2019.08.002
[3]	付强, 陈向阳, 郑子亮, 李擎, 贺威. 仿生扑翼飞行器的视觉感知系统研究进展. 工程科学学报, 2019, 41(12): 1512−1519 Fu Qiang, Chen Xiang-Yang, Zheng Zi-Liang, Li Qing, He Wei. Research progress on visual perception system of bionic flapping-wing aerial vehicles. Chinese Journal of Engineering, 2019, 41(12): 1512−1519
[4]	Gerdes J, Holness A, Perez-Rosado A, Roberts L, Greisinger A, Barnett E, et al. Robo Raven: a flapping-wing air vehicle with highly compliant and independently controlled wings. Soft Robotics, 2014, 1(4): 275−288 doi: 10.1089/soro.2014.0019
[5]	Keennon M, Klingebiel K, Won H. Development of the nano hummingbird: A tailless flapping wing micro air vehicle. 50th AIAA aerospace sciences meeting including the new horizons forum and aerospace exposition, 2012588
[6]	Chen Y F, Wang H Q, Helbling E F, Jafferis N T, Zufferey R, Ong A, et al. A biologically inspired flapping-wing hybrid aerial–aquatic microrobot. Science Robotics, 2017, 2(11): 5619 doi: 10.1126/scirobotics.aao5619
[7]	Fu Q, Wang J, Gong L, Wang J Y, He W. Obstacle avoidance of flapping-wing air vehicle based on optical flow and fuzzy control. Transactions of Nanjing University of Aeronautics and Astronautics, 2021, 38(2): 206−215
[8]	Banazadeh A, Taymourtash N. Adaptive attitude and position control of an insect-like flapping-wing air vehicle. Nonlinear Dynamics, 2016, 85(1): 47−66 doi: 10.1007/s11071-016-2666-8
[9]	Qian C, Fang Y C, Li Y P. Neural network-based hybrid three-dimensional position control for a flapping wing aerial vehicle. IEEE Transactions on Cybernetics, 2022, 53(10): 6095−6108 doi: 10.1109/tcyb.2022.3166566
[10]	Huang H F, Liu Y T, Niu T, Wang C, Zou Y, He W. Vector field path following for a micro flapping-wing robot. Science China. Information Sciences, 2024, 67(6): 164201 doi: 10.1007/s11432-023-4027-x
[11]	贺威, 刘上平, 黄海丰, 付强, 孙长银. 独立驱动的仿鸟扑翼飞行机器人的系统设计与实验. 控制理论与应用, 2022, 39(1): 12−22 doi: 10.7641/CTA.2021.00394 He Wei, Liu Shang-Ping, Huang Hai-Feng, Fu Qiang, Sun Chang-Yin. System design and experiment of an independently driven bird-like flapping-wing robot. Control Theory & Applications, 2022, 39(1): 12−22 doi: 10.7641/CTA.2021.00394
[12]	Wen G X, Yu D X, Zhao Y L. Optimized fuzzy attitude control of quadrotor unmanned aerial vehicle using adaptive reinforcement learning strategy. IEEE Transactions on Aerospace and Electronic Systems, 2024, 60(5): 6075−6083 doi: 10.1109/TAES.2024.3401668
[13]	Liu M, Ma D X, Li S. Neural dynamics for adaptive attitude tracking control of a flapping wing micro aerial vehicle. Neurocomputing, 2021, 456: 364−372 doi: 10.1016/j.neucom.2021.05.088
[14]	Wang L, Jiang W Y, Zhao L F, Jiao Z X. Modeling and hover flight control of a micromechanical flapping-wing aircraft inspired by wing $–$tail interaction. IEEE/ASME Transactions on Mechatronics, 2023, 28(6): 3132−3142 doi: 10.1109/TMECH.2023.3254921
[15]	Tijmons S, De Croon G C H E, Remes B D W, De Wagter C, Mulder M. Obstacle avoidance strategy using onboard stereo vision on a flapping wing MAV. IEEE Transactions on Robotics, 2017, 33(4): 858−874 doi: 10.1109/TRO.2017.2683530
[16]	付强, 张树禹, 王久斌, 冯富森. 基于外部单目视觉的仿生扑翼飞行器室内定高控制. 工程科学学报, 202, 42(2): 249-256 Fu Qiang, Zhang Shu-Yu, Wang Jiu-Bin, Feng Fu-Sen. Indoor fixed-height control for bio-inspired flapping-wing aerial vehicles based on off-board monocular vision. Chinese Journal of Engineering, 2020, 42(2): 249-256
[17]	付强, 张祥, 赵民, 张春华, 贺威. 仿生扑翼飞行器风洞实验研究进展. 工程科学学报, 2022, 44(4): 767−779 Fu Qiang, Zhang Xiang, Zhao Min, Zhang Chun-Hua, He Wei. Research progress on the wind tunnel experiment of a bionic flapping-wing aerial vehicle. Chinese Journal of Engineering, 2022, 44(4): 767−779
[18]	Lv Z F, Xiao L, Du Y S, Niu G H, Xing C W, Xu W Y. Multi-agent reinforcement learning based UAV swarm communications against jamming. IEEE Transactions on Wireless Communications, 2023, 22(12): 9063−9075 doi: 10.1109/TWC.2023.3268082
[19]	Kong L H, He W, Chen W S, Zhang H, Wang Y N. Dynamic movement primitives based robot skills learning. Machine Intelligence Research, 2023, 20(3): 396−407 doi: 10.1007/s11633-022-1346-z
[20]	Zhang C X, Lu W J, Zhao S Y, Wu J, Zhu X Y, Liu Z J, et al. Enhancing Attitude Tracking with Self-learning Control using Tanh-type Learning Intensity. IEEE Transactions on Automation Science and Engineering, 2025, 22: 16976−16986 doi: 10.1109/TASE.2025.3581953
[21]	胡海天. 基于强化学习的微型扑翼飞行器机动动作控制研究[硕士学位论文], 哈尔滨工业大学, 中国, 2025 Hu Hai-Tian. Research on maneuver control of flapping wing micro aerial vehicle based on reinforcement learning. [Master thesis], Harbin Institute of Technology, China, 2025
[22]	Sun M, Tang J. Unsteady aerodynamic force generation by a model fruit fly wing. Journal of Experimental Biology, 2002, 205(1): 55−70 doi: 10.1242/jeb.205.1.55
[23]	Zhang C X, Ahn C K, Wu J, He W. Online-learning control with weakened saturation response to attitude tracking: A variable learning intensity approach. Aerospace Science and Technology, 2021, 117: 106981 doi: 10.1016/j.ast.2021.106981
[24]	Zhang C X, Xiao B, Wu J, Li B. On low-complexity control design to spacecraft attitude stabilization: An online-learning approach. Aerospace Science and Technology, 2021, 110: 106441 doi: 10.1016/j.ast.2020.106441
[25]	张良, 何修宇, 王者奇, 贺威. 扑翼飞行器的变幅飞行策略的设计与实现. 机器人, 2025, 47(1): 76−84 Zhang Liang, He Xiu-Yu, Wang Zhe-Qi, He Wei. Design and Implementation of a Variable-amplitude Flight Strategy for Flapping-wing Aircraft. Robot, 2025, 47(1): 76−84