一种小型四轴飞行器目标跟踪控制算法

doi:10.19665/j.issn1001-2400.2021.05.015

摘要/Abstract

摘要：

为了提高利用四轴飞行器单目摄像头传回的图片帧执行目标跟踪任务时的实时性和稳定性,提出一种基于注意力模型的图片帧预处理和基于反馈控制的四轴飞行器姿态调整的算法。首先,通过注意力模型对四轴飞行器采集到的视频图像进行预处理,选出对跟踪结果贡献度较大的部分帧;其次,将图像预处理阶段选出的图片帧进一步处理,得到选框像素点的位置偏差和面积偏差,记作反馈控制的误差;最后,通过反馈控制,将该误差与四轴飞行器的电机转速相转化,并将电机转速控制指令发送给飞行器,使其及时作出偏航和前后移动的姿态调整。通过在特洛(Tello)飞行器平台上的实际飞行验证,实验结果表明,所提出的目标跟踪算法在图像处理实时性和飞行器飞行稳定性两方面都有提升。所提出的算法可为四轴飞行器图像处理应用提供有力的技术支持。

关键词: 四轴飞行器, 目标跟踪, 反馈控制, 注意力模型, 姿态调整

Abstract:

In order to improve the real-time performance and stability of target tracking by using the video frames transmitted from the monocular camera of a quad-rotor,an algorithm for video frame preprocessing based on the attention model and attitude adjustment of the quad-rotor based on feedback control is proposed.First,the attention model is used to preprocess the video images collected by the quad-rotor,and some frames which contribute more to the tracking results are selected.Second,the error of feedback control is obtained by calculating the position deviation and area deviation of the pixel points of the marquee.Finally,the error of feedback control is converted into the numerical change in motor speed of the four-axis aircraft and the motor speed control instruction are sent to the aircraft,which can make attitude adjustment of the yaw and forward and backward movement in time.The actual flight verification is executed on the Tello aircraft platform,andexperimental results shows that the target tracking algorithm proposed in this paper improves both the real-time performance of image processing and the flight stability of aircraft.Therefore,the algorithm proposed in this paper can provide a powerful technical support for the image processing application of quad-rotor aircraft.

Key words: quad-rotor helicopter, target tracking, feedback control, attention model, attitude adjustment

中图分类号:

TP391.41

顾兆军,陈辉,王家亮,高冰. 一种小型四轴飞行器目标跟踪控制算法[J]. 西安电子科技大学学报, 2021, 48(5): 117-127.

GU Zhaojun,CHEN Hui,WANG Jialiang,GAO Bing. Target tracking control algorithm for small size quad-rotor helicopter[J]. Journal of Xidian University, 2021, 48(5): 117-127.

图/表 19

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

图11

图12

图13

图14

图15

图16

图17

表1

表2

参考文献 22

[1]	RADOGLOU-GRAMMATIKIS P, SARIGIANNIDIS P, LAGKAS T, et al. A Compilation of UAV Applications for Precision Agriculture[J]. Computer Networks, 2020, 172:107148. doi: 10.1016/j.comnet.2020.107148
[2]	WANG J, HAN L, LI X, et al. Time-Varying Formation of Second-Order Discrete-Time Multi-Agent Systems Under Nonuniform Communication Delays and Switching Topology with Application to UAV Formation Flying[J]. IET Control Theory and Applications, 2020, 14(14):1947-1956. doi: 10.1049/cth2.v14.14
[3]	GUO Q, ZHU Y, TANG Y, et al. CFD Simulation and Experimental Verification of the Spatial and Temporal Distributions of the Downwash Airflow of a Quad-Rotor Agricultural UAV in Hover[J]. Computers and Electronics in Agriculture, 2020, 172:105343. doi: 10.1016/j.compag.2020.105343
[4]	LIU L, PAN F, AN Z, et al. Target Tracking Control and Semi-Physical Simulation of Qball-X4 Quad-Rotor Unmanned Aerial Vehicle[J]. International Journal of Advanced Robotic Systems, 2017, 14(1):172988141668695.
[5]	LIAO S, ZHU R, WU N, et al. Path Planning for Moving Target Tracking by Fixed-Wing UAV[J]. Defence Technology, 2020, 16(4):811-824. doi: 10.1016/j.dt.2019.10.010
[6]	KIM J, SHIM D H. A Vision-Based Target Tracking Control System of a Quadrotor by Using a Tablet Computer[C]// International Conference on Unmanned Aircraft Systems.Piscataway:IEEE, 2013:1165-1172.
[7]	TAO Y, XIE G, CHEN Y, et al. A PID and Fuzzy Logic Based Method for Quadrotor Aircraft Control Motion[J]. Journal of Intelligent & Fuzzy Systems 2016, 31:2975-2983.
[8]	HAAG K, DEITSCH S, GALLWITZ F. Correlation Filter Based Visual Trackers for Person Pursuit Using a Low-Cost Quadrotor[C]// International Conference on Innovations for Community Services.Piscataway:IEEE, 2015:1-8.
[9]	ZHANG H, SUN M, LI Q, et al. An Empirical Study of Multi-Scale Object Detection in High Resolution UAV Images[J]. Neurocomputing, 2021, 421:173-182. doi: 10.1016/j.neucom.2020.08.074
[10]	ZHANG R, SHAO Z, HUANG X, et al. Object Detection in UAV Images via Global Density Fused Convolutional Network[J]. Remote Sensing, 2020, 12(19):3140. doi: 10.3390/rs12193140
[11]	李冬霞, 李春鸣, 赵文强, 等. 无人机中继广播通信系统航迹优化方法[J]. 西安电子科技大学学报, 2018, 45(3):149-154.
	LI Dongxia, LI Chunming, ZHAO Wenqiang, et al. Trajectory Optimization Method for the UAV Relaying Broadcast Communication System[J]. Journal of Xidian University, 2018, 45(3):149-154.
[12]	HUANG F, WU P, LI X. Distributed Flocking Control of Quad-Rotor UAVs with Obstacle Avoidance Under the Parallel-Triggered Scheme[J]. International Journal of Control,Automation and Systems, 2021, 19(3):1375-1383. doi: 10.1007/s12555-019-0315-y
[13]	DAI X, MAO Y, HUANG T, et al. Automatic Obstacle Avoidance of Quadrotor UAV via CNN-Based Learning[J]. Neurocomputing, 2020, 402:346-358. doi: 10.1016/j.neucom.2020.04.020
[14]	王术波, 韩宇, 陈建, 等. 基于ADRC迭代学习控制的四旋翼无人机姿态控制[J]. 航空学报, 2020, 41(12):319-331.
	WANG Shubo, HAN Yu, CHEN Jian, et al. UAV Attitude Active Disturbance Rejection Control Based on Iterative Learning Control[J]. Acta Aeronautica et Astronautica Sinica, 2020, 41(12):319-331.
[15]	马卓然, 马建峰, 苗银宾, 等. 无人机网络中基于状态迁移的访问控制模型[J]. 西安电子科技大学学报, 2018, 45(6):44-50.
	MA Zhuoran, MA Jianfeng, MIAO Yinbin, et al. StateTransition-Based Access Control Model in the UAV Network[J] Journal of Xidian University, 2018, 45(6):44-50.
[16]	GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation[C]// Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition.Piscataway:IEEE, 2014:580-587.
[17]	LIU W, ANGUELOV D, ERHAN D, et al. SSD:Single Shot MultiBox Detector[C]// European Conference on Computer Vision.Heidelberg:Springer, 2016:21-37.
[18]	REDMON J, DIVVALA S, GIRSHICK R, et al. You Only Look Once:Unified,Real-Time Object Detection[C]// Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition.Piscataway:IEEE, 2016:779-788.
[19]	HENRIQUES J F, CASEIRO R, MARTINS P, et al. High-Speed Tracking with Kernelized Correlation Filters[J]. IEEE Transactions on Pattern Analysisand Machine Intelligence, 2015, 37(3):583-596.
[20]	LI Y, ZHU J. A Scale Adaptive Kernel Correlation Filter Tracker with Feature Integration[C]// European Conference on Computer Vision.Heidelberg:Springer, 2014:254-265.
[21]	WU Y, LIM J, YANG M H. Online Object Tracking:A Benchmark[C]// 2013 IEEE Computer Vision and Pattern Recognition.Piscataway:IEEE, 2013:2411-2418.
[22]	PAN F, LIU L, XUE D. Optimal PID Controller Design with Kalman Filter for Qball-X4 Quad-Rotor Unmanned Aerial Vehicle[J]. Transactions of the Institute of Measurement and Control, 2016, 39(12):1785-1797. doi: 10.1177/0142331216656753

模型	处理图片帧数量	单位处理时间/s	总处理时间/s
YOLOV4+KCF模型(CPU)	50	0.326	16.30
YOLOV4+KCF-tiny模型(CPU)	50	0.273	13.65
YOLOV4+KCF-tiny模型(GPU)	50	0.015	0.75
文中基于注意力模型的目标检测跟踪算法(GPU)	6	0.015	0.09