基于飞行数据的无人机平飞动作质量评价模型

滕怀亮; 李本威; 高永; 杨栋; 张赟

doi:10.13700/j.bh.1001-5965.2019.0029

基于飞行数据的无人机平飞动作质量评价模型

doi: 10.13700/j.bh.1001-5965.2019.0029

滕怀亮¹,
李本威¹,
高永^1, ,,
杨栋²,
张赟¹

1.
海军航空大学航空基础学院, 烟台 264001
2.
中国人民解放军92074部队, 宁波 315000

基金项目:

国家自然科学基金 51505492

山东省泰山学者建设工程专项

详细信息

作者简介:
滕怀亮  男, 硕士研究生。主要研究方向:航空发动机状态监控与健康评估技术

李本威  男, 博士, 教授, 博士生导师。主要研究方向:航空宇航推进理论与工程

高永  男, 博士, 副教授, 硕士生导师。主要研究方向:飞行器设计

通讯作者:
高永, E-mail:gaoyongbh@sina.com

中图分类号: V279
计量
- 文章访问数: 1039
- HTML全文浏览量: 102
- PDF下载量: 520
- 被引次数: 0
出版历程
- 收稿日期: 2019-01-22
- 录用日期: 2019-04-15
- 网络出版日期: 2019-10-20

Quality evaluation model of unmanned aerial vehicle's horizontal flight maneuver based on flight data

1.
Aviation Foundation College, Naval Aviation University, Yantai 264001, China
2.
Troop 92074 of People's Liberation Army, Ningbo 315000, China

Funds:

National Natural Science Foundation of China 51505492

Taishan Scholar Construction Foundation of Shandong Province

More Information

Corresponding author: GAO Yong, E-mail:gaoyongbh@sina.com

摘要

摘要:
针对无人机（UAV）操纵质量高低多依赖专家主观评判、不同飞行动作缺乏针对性评判标准的问题，建立了根据飞行数据客观评价无人机平飞动作质量的模型。首先通过建立平飞判别规则识别无人机飞行数据中的平飞动作数据段，然后根据布林通道理论计算各平飞数据段内多个飞行参数的得分值，最后通过"熵权法"确定各参数的权重，进而得到反映不同无人机操控手平飞动作质量的指标数据。将一次四边平飞训练任务中4组不同无人机操控手操纵和1组无人机自主控制的飞行数据输入该模型，计算结果表明，模型可以较好地识别平飞动作并区分不同操控手平飞动作质量的高低，可以为无人机操控手的训练提供参考。
- 质量评价 /
- 无人机 /
- 平飞动作 /
- 布林通道 /
- 熵权法
Abstract:
The unmanned aerial vehicle (UAV) manipulator's flying performance mostly relies on experts' subjective evaluation and different flight maneuvers lack pertinent evaluation criteria, so a model which uses flight data to evaluate the horizontal flight quality of UAV is established. Firstly, the flight data segment of UAV's horizontal flight maneuver was identified by the flight discrimination rules. Then, according to Bollinger bands theory, the scores of multiple flight parameters in each flight data segment were calculated. Finally, the weight of each parameter was determined by the entropy weight method and the indexes reflecting the horizontal flight maneuver's quality of different UAV manipulators were obtained. In a quadrilateral flying training mission, four groups of different UAV maneuvers' flight data and one group of flight data under autonomous control were input into the model. The calculation results show that the model can well identify the horizontal flight maneuver and distinguish the horizontal flight maneuver's quality of different manipulators, which can provide advice for the training of UAV manipulators.
- quality evaluation /
- unmanned aerial vehicle /
- horizontal flight maneuver /
- Bollinger bands /
- entropy weight method

HTML全文

图 1 评价模型实现流程图

Figure 1. Flowchart for realizing evaluation model

下载: 全尺寸图片幻灯片

图 2 典型平飞数据段

Figure 2. Typical horizontal flight data segment

下载: 全尺寸图片幻灯片

图 3 布林线构成的数据带

Figure 3. Data band constituted by Bollinger lines

下载: 全尺寸图片幻灯片

图 4 无人机航迹图

Figure 4. UAV path map

下载: 全尺寸图片幻灯片

图 5 高度随时间的变化

Figure 5. Variation of height with time

下载: 全尺寸图片幻灯片

图 6 高度变化率插值结果

Figure 6. Interpolation results of height change rate

下载: 全尺寸图片幻灯片

图 7 操控手1得分

Figure 7. Scores of Manipulator 1

下载: 全尺寸图片幻灯片

图 8 自主控制模式下的平飞数据段

Figure 8. Horizontal flight data segments in autonomous control mode

下载: 全尺寸图片幻灯片

表 1 无人机操控手飞行参数评分

Table 1. Scoring for flight parameters of UAV manipulators

平飞段	参数1	参数2	…	参数n
1	X₁₁	X₁₂	…	X_1n
2	X₂₁	X₂₂	…	X_2n
⋮	⋮	⋮	⋮	⋮
m	X_m1	X_m2	…	X_mn

下载: 导出CSV

表 2 平飞数据段

Table 2. Horizontal flight data segments

序号	起始时间/s	结束时间/s
1	4 666	4 688
2	4 698	4 721
3	4 757	4 783
4	4 849	4 872
5	5 428	5 457
6	5 787	5 808
7	6 172	6 206
8	6 366	6 393
9	6 674	6 699
10	6 787	6 808
11	6 871	6 892
12	7 045	7 074
13	7 237	7 269
14	8 089	8 114
15	8 186	8 210
16	8 266	8 300
17	8 428	8 458
18	8 662	8 698
19	8 852	8 888

下载: 导出CSV

表 3 评分表

Table 3. Score sheet

序号	高度得分	速度得分	俯仰得分	偏航得分	滚转得分
1	1.400 4	0.627 5	0.202 2	0.220 6	0.901 7
2	4.208 2	0.672 7	0.339 7	0.275 3	1.118 8
3	1.737 1	0.819 9	0.332 7	0.226 7	0.859
4	1.185 9	0.684 1	0.066 5	0.147 9	0.515 1
5	1.307 8	0.557 7	0.281 8	0.213 3	0.747 4
6	1.006 2	0.450 3	0.120 5	0.183	0.698 8
7	1.735 8	0.784 9	0.132 9	0.197 4	0.746 8
8	4.139 5	0.564	0.152 1	0.165 5	0.674 2
9	1.528 7	0.422 2	0.086 1	0.152 7	0.527
10	1.49 2	0.462 8	0.174	0.232 8	0.773 3
11	1.009 7	0.445 4	0.059 3	0.163 6	0.540 6
12	0.999 9	0.371 8	0.130 1	0.197 4	0.704 5
13	1.951 1	0.465 5	0.148 8	0.201 5	0.680 9
14	2.379 8	0.641 2	0.065 9	0.152 3	0.483
15	1.930 5	1.171 6	0.301 9	0.367 6	1.366 6
16	1.957 7	0.594 1	0.157 8	0.191 6	0.808 9
17	3.418 5	0.724 9	0.123 6	0.209 6	1.010 2
18	0.876 7	0.595 5	0.213	0.252 8	0.955 1
19	0.653 1	0.556 6	0.118 5	0.165 8	0.646 3

下载: 导出CSV

表 4 最终得分排名

Table 4. Final score ranking

操控主体	得分	平飞段数	排名
操控手1	0.208 07	19	第4名
操控手2	0.170 01	28	第3名
操控手3	0.145 02	39	第2名
操控手4	0.351 29	5	第5名
自主操纵	0.100 78	49	第1名

下载: 导出CSV

参考文献(18)

[1]	DAS S, MATTHEWS B L, LAWRENCE R. Fleet level anomaly detection of aviation safety data[C]//IEEE Conference on Prognostics and Health Management.Piscataway, NJ: IEEE Press, 2011: 1-10.
[2]	ZHOU K, WANG L X, TAN X S.Flying qualities reduction of fly-by-wire commercial aircraft with reconfiguration flight control laws[J].Procedia Engineering, 2011, 17:179-196. doi: 10.1016/j.proeng.2011.10.021
[3]	WANG L, WU C X, SUN R S.An analysis of flight quick access recorder(QAR) data and its applications in preventing landing incidents[J].Reliability Engineering and System Safety, 2014, 127:86-96. doi: 10.1016/j.ress.2014.03.013
[4]	张玉叶, 王颖颖, 王春歆, 等.分析参数相关和时序特征的飞行动作识别方法[J].计算机工程与应用, 2016, 52(5):246-249. doi: 10.3778/j.issn.1002-8331.1403-0283 ZHANG Y Y, WANG Y Y, WANG C X, et al.Acrobatic maneuver reorganization method compared with parameters relevance and feature of sequence change[J].Computer Engineering and Applications, 2016, 52(5):246-249(in Chinese). doi: 10.3778/j.issn.1002-8331.1403-0283
[5]	沈一超, 倪世宏, 张鹏.基于贝叶斯网络的飞行动作识别方法[J].计算机工程与应用, 2017, 53(24):161-167. doi: 10.3778/j.issn.1002-8331.1707-0156 SHEN Y C, NI S H, ZHANG P.Flight action recognition method based on Bayesian network[J].Computer Engineering and Applications, 2017, 53(24):161-167(in Chinese). doi: 10.3778/j.issn.1002-8331.1707-0156
[6]	HOLMBERG J, LEONARD J, KING D, et al.Flying qualities specifications and design standards for unmanned air vehicles[C]//AIAA Atmospheric Flight Mechanics Conference and Exhibit.Reston, VA: AIAA, 2008: 1-14.
[7]	COTTING C M.An initial study to categorize unmanned aerial vehicles for flying qualities evaluation[C]//AIAA Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace Exposition.Reston, VA: AIAA, 2009: 1-12.
[8]	刘莉雯, 张天伟, 茹斌.多参数融合的飞行品质评估模型的建立[J].计算机工程与科学, 2016, 38(6):1262-1268. doi: 10.3969/j.issn.1007-130X.2016.06.030 LIU L W, ZHANG T W, RU B.A flying qualities assessment model based on multi-parameter integration[J].Computer Engineering and Science, 2016, 38(6):1262-1268(in Chinese). doi: 10.3969/j.issn.1007-130X.2016.06.030
[9]	张龙, 徐开俊.采用模糊逻辑理论的飞行操纵品质评估模型[J].中国民航飞行学院学报, 2016, 27(5):9-13. doi: 10.3969/j.issn.1009-4288.2016.05.002 ZHANG L, XU K J.Flight operation quality assessment model using fuzzy logic theory[J].Journal of Civil Aviation Flight University of China, 2016, 27(5):9-13(in Chinese). doi: 10.3969/j.issn.1009-4288.2016.05.002
[10]	柳忠起, 袁修干, 樊瑜波.基于BP神经网络的飞行绩效评价模型[J].北京航空航天大学学报, 2010, 36(4):403-406. https://bhxb.buaa.edu.cn/CN/abstract/abstract8467.shtml LIU Z Q, YUAN X G, FAN Y B.Pilot performance evaluation model based on BP neural networks[J].Journal of Beijing University of Aeronautics and Astronautics, 2010, 36(4):403-406(in Chinese). https://bhxb.buaa.edu.cn/CN/abstract/abstract8467.shtml
[11]	夏长俊, 孙校书, 李冰.基于飞行数据的无人机平飞质量分析方法[J].海军航空工程学院学报, 2017, 32(5):478-482. http://d.old.wanfangdata.com.cn/Periodical/hjhkgcxyxb201705012 XIA C J, SUN X S, LI B.Analysis method of UAV level flight quality based on flight data[J].Journal of Naval Aeronautical and Astronautical University, 2017, 32(5):478-482(in Chinese). http://d.old.wanfangdata.com.cn/Periodical/hjhkgcxyxb201705012
[12]	彭阳, 彭晓明.一种无人机飞行动作质量分析方法[J].飞行力学, 2017, 35(6):65-69. doi: 10.3969/j.issn.1002-0853.2017.06.015 PENG Y, PENG X M.A flight maneuver quality evaluation method for UAVs[J].Flight Dynamics, 2017, 35(6):65-69(in Chinese). doi: 10.3969/j.issn.1002-0853.2017.06.015
[13]	ZHOU S H, QIAN S L, QIAO X D, et al.The UAV landing quality evaluation research based on combined weight and VIKOR algorithm[J].Wireless Personal Communications, 2018, 102(2):2047-2062. doi: 10.1007/s11277-018-5254-z
[14]	杨俊, 谢寿生.基于模糊支持向量机的飞机飞行动作识别[J].航空学报, 2005, 26(6):738-742. doi: 10.3321/j.issn:1000-6893.2005.06.016 YANG J, XIE S S.Fuzzy support vector machines based recognition for aero-plane flight action[J].Acta Aeronautica et Astronautica Sinica, 2005, 26(6):738-742(in Chinese). doi: 10.3321/j.issn:1000-6893.2005.06.016
[15]	SHARP C S, SHAKERNIA O, SASTRY S S.A vision system for landing an unmanned aerial vehicle[C]//Proceedings of the 2001 IEEE International Conference on Robotics & Automation.Piscataway, NJ: IEEE Press, 2001: 1720-1727.
[16]	李樾, 陈清阳, 侯中喜.自适应引导长度的无人机航迹跟踪方法[J].北京航空航天大学学报, 2017, 43(7):1481-1490. https://bhxb.buaa.edu.cn/CN/abstract/abstract14109.shtml LI Y, CHEN Q Y, HOU Z X.Path following method with adaptive guidance length for unmanned aerial vehicles[J].Journal of Beijing University of Aeronautics and Astronautics, 2017, 43(7):1481-1490(in Chinese). https://bhxb.buaa.edu.cn/CN/abstract/abstract14109.shtml
[17]	LIU W, ZHENG W A.Stochastic volatility model and technical analysis of stock price[J].Acta Mathematica Sinica, English Series, 2011, 27(7):1283-1296. doi: 10.1007/s10114-011-9468-1
[18]	王玉伟, 高永.基于稳健回归算法的无人机数据预处理技术研究[J].舰船电子工程, 2018, 38(11):38-41. doi: 10.3969/j.issn.1672-9730.2018.11.011 WANG Y W, GAO Y.Research on UAV data preprocessing technology based on robust regression algorithm[J].Ship Electronic Engineering, 2018, 38(11):38-41(in Chinese). doi: 10.3969/j.issn.1672-9730.2018.11.011