Research on roll angle estimation method based on deep learning
-
摘要: 姿态测量技术是载体运动状态和安全监测的基础. 载体自旋运动使得飞行器各姿态角之间互相耦合，对载体的飞行控制带来严重影响. 针对载体滚转下GNSS信号入射方向周期性变化特征，本文提出一种长短期记忆神经网络的深度学习方法，以确定载体的实时滚转角. 通过对载体滚转状态下单天线接收卫星信号能量特征的分析，得到载体实时滚转角与接收信号能量幅值关联变化模型，并分析了卫星在轨运行时其空间位置改变对该模型的影响；然后采用长短期记忆(long short termmemory, LSTM)神经网络方法对实测信号中的周期性变化特征进行训练，得到网络各项参数；最后将训练参数用于对实时接收的信号能量进行预测及降噪，并将预测结果通过模型匹配进行载体实时滚转角测算. 为验证文中所提出方法的性能，开展了对天滚转实验. 实验结果表明：LSTM深度学习方法可还原复杂的信号能量特征，并实现实时载体滚转角测算.
-
关键词:
- GNSS /
- 深度学习 /
- 长短期记忆(LSTM)神经网络 /
- 滚转角 /
- 天线增益
Abstract: Attitude measurement technology serves as a fundamental component in monitoring vehicle motion states and ensuring safety. The spinning motion of vehicles leads to a coupling of attitude angles, which significantly impacts flight control. In this paper, a deep learning approach based on the Long Short-Term Memory (LSTM) neural network is proposed to determine the real-time roll angle of a vehicle. The energy characteristics exhibited by a single antenna receiving satellite signals during the vehicle's rolling state have been analyzed in detail. A correlation between the real-time roll angle and the energy amplitude of the received signal has been established. The influence of changing satellite positions on these measurements is also discussed. Subsequently, the LSTM neural network training method is employed to extract periodic variation features from the measured signals, thereby obtaining various network parameters. These parameters are then used to predict and denoise the received signal, with the roll angle being computed by model matching. To validate the efficacy of the proposed method, a rolling experiment was conducted. The experimental results demonstrate that the LSTM-based deep learning approach effectively restores the features of the received signals, enabling accurate real-time measurement of the vehicle's roll angle.-
Key words:
- GNSS /
- deep learning /
- LSTM /
- roll angle /
- antenna gain
-
表 1 不同方法下滚转角估计误差分析
算法 2 r/s
标准差/(°)10 r/s
标准差/(°)20 r/s
标准差/(°)平均滚转角
估计标准差/(°)LSTM 7.703 6.937 10.300 8.313 CNN-
LSTM13.017 13.525 19.559 15.367 LS 12.780 14.227 19.371 15.460 
下载: 导出CSV
-
[1] LE V D, NGUYEN A T, NGUYEN L H, et al. Effectiveness analysis of spin motion in reducing dispersion of sounding rocket flight due to thrust misalignment[J]. International journal of aeronautical and space sciences, 2021, 22(5): 1194-1208. DOI: 10.1007/s42405-021-00383-x [2] YANG D, XIONG Y L, REN Q, et al. Nutation instability of spinning solid rocket motor spacecraft[J]. Chinese journal of aeronautics, 2017, 30(4): 1363-1372. DOI: 10.1016/j.cja.2017.06.005 [3] YANG W C, WANG Z Q, SHEN C W, et al. Design of a roll angle measuring sensor[J]. IEEE access, 2020(8): 115159-115166. DOI: 10.1109/access.2020.3004365 [4] GROVES P. Principles of GNSS, inertial, and multisensor integrated navigation systems[M]. Artech, 2007: 1. [5] AL-RAWASHDEH Y M, ELSHAFEI M, EL-FEIRIK S. Passive attitude estimation using gyroscopes and all-accelerometer IMU[C]// 2016 7th International Conference on Mechanical and Aerospace Engineering (ICMAE). IEEE, 2016. DOI: 10.1109/icmae.2016.7549568. [6] 代桃高, 宫帅帅, 魏明, 等. 一种GNSS双天线姿态确定及点位标定方法研究[J]. 全球定位系统, 2019, 44(6): 110-115. DOI: 10.13442/j.gnss.1008-9288.2019.06.018 [7] 陆尤明, 刘刚, 崔晓伟, 等. 基于双天线联合的GNSS信号抗旋转跟踪算法[J]. 清华大学学报(自然科学版), 2021, 61(9): 1015-1024. DOI: 10.16511/j.cnki.qhdxxb.2021.21.025 [8] 廉璞, 牟东, 青泽, 等. 侵彻弹药姿态测量技术研究现状及发展[J]. 探测与控制学报, 2021, 43(2): 1-9. [9] NGUYEN D N, NGUYEN T A. Investigate the relationship between the vehicle roll angle and other factors when steering[J]. Modelling and simulation in engineering, 2023, 2023: 1-15. DOI: 10.1155/2023/6069078 [10] COHEN C, PARKINSON B, MCNALLY B. Flight tests of attitude determination using GPS compared against an inertial navigation unit[J]. Navigation, 1994, 41(1): 83-97. DOI: 10.1002/j.2161-4296.1994.tb02323.x [11] LI N, ZHAO L, LI L, et al. Integrity monitoring of high-accuracy GNSS-based attitude determination[J]. GPS solutions, 2018, 22(4): 1-13. DOI: 10.1007/s10291-018-0787-x [12] IM H C, LEE S J. GPS signal tracking on a multi-antenna mounted spinning vehicle by compensating for the spin effects[J]. International journal of control, automation and systems, 2018, 16(2): 867-874. DOI: 10.1007/s12555-016-0705-3 [13] ZHARKOV M V, VEREMEENKO K K, ANTONOV D A, et al. Attitude determination using ambiguous GNSS phase measurements and absolute angular rate measurements[J]. Gyroscopy and navigation (Online), 2018, 9(4): 277-286. DOI: 10.1134/s2075108718040090 [14] DOTY J H, ANDERSON D A, BYBEE T. A demonstration of advanced spinning-vehicle navigation[C]. In Proceedings of the ION NTM, 2004, San Diego, CA, USA. [15] KIM J W, KANG H W, HWANG D H, et al. Signal tracking method of GNSS receivers for spinning vehicles[J]. International journal of control, automation and systems, 2012, 10(3): 529-535. DOI: 10.1007/s12555-012-0309-5 [16] BAHDER T B. Attitude determination from single-antenna carrier-phase measurements[J]. Journal of applied physics, 2002, 91(7): 4677-4684. DOI: 10.1063/1.1448871 [17] DENG Z L, SHEN Q, DENG Z W. Roll angle measurement for a spinning vehicle based on GPS signals received by a single-patch antenna[J]. Sensors, 2018, 18(10): 3479. DOI: 10.3390/s18103479 [18] LIU Y, LI H, DU X. Roll attitude measurement technique based on GPS signal power[C]. 2019 IEEE International Conference on Unmanned Systems (ICUS), Beijing, China, 2019. [19] HEATON J, GOODFELLOW I, BENGIO Y, et al. Deep learning[J]. Genetic programming and evolvable machines, 2018, 19(1): 305-307. DOI: 10.1007/s10710-017-9314-z [20] GRAVES A, GREG W, MALCOLM R, et al. Hybrid computing using a neural network with dynamic external memory[J]. Nature, 2016, 538(7626): 471-476. DOI: 10.1038/nature20101 [21] BROSSARD M, BARRAU A, BONNABEL S. AI-IMU dead-reckoning[J]. IEEE transactions on intelligent vehicles, 2020, 5(4): 585-595. DOI: 10.1109/TIV.2020.2980758 [22] TANG H, NIU X, ZHANG T, et al. OdoNet: untethered speed aiding for vehicle navigation without hardware wheeled odometer[J]. IEEE sensors journal, 2022, 22(12): 12197-12208. DOI: 10.1109/JSEN.2022.3169549 [23] DUBEY A K, KUMAR A, VICENTE G D, et al. Study and analysis of SARIMA and LSTM in forecasting time series data[J]. Sustainable energy technologies and assessments, 2021, 47: 101474. DOI: 10.1016/j.seta.2021.101474 [24] LIU Y, XU B, CHEN Z K, et al. Detection and mitigation of time synchronization attacks based on long short-term memory neural network[J]. GPS solutions, 2023, 28(1): 46. DOI: 10.1007/s10291-023-01587-2 [25] 尹玲, 尹京苑, 孙宪坤, 等. 缺失GPS时间序列的神经网络补全[J]. 测绘科学技术学报, 2018, 35(4): 331-336. DOI: CNKI:SUN:JFJC.0.2018-04-001. [26] CHEN Q J, ZHANG Q, NIU X J. Estimate the pitch and heading mounting angles of the IMU for land vehicular GNSS/INS integrated system[J]. IEEE transactions on intelligent transportation systems, 2021, 22(10): 6503-6515. DOI: 10.1109/TITS.2020.2993052 [27] SRIVASTAVA N, HINTON G, KRIZHEVSKY A, et al. Dropout: a simple way to prevent neural networks from overfitting[J]. Journal of machine learning research, 2014, 15(1): 1929-1958. DOI: 10.5555/2627435.2670313 [28] ARORA S, COHEN N, GOLOWICH N, et al. A convergence analysis of gradient descent for deep linear neural networks[C]. 7th International Conference on Learning Representations, 2019, New Orleans, USA. [29] DU S S, LEE J D, LI H C, et al. Gradient descent finds global minima of deep neural networks[C]. 36th International Conference on Machine Learning, 2019, Long Beach, USA. [30] DU S S, ZHAI X Y, POCZOS B, et al. Gradient descent provably optimizes over-parameterized neural networks[C]. 7th International Conference on Learning Representations, 2019, New Orleans, USA. -
点击查看大图
图(11) / 表(1)
计量
- 文章访问数: 22
- HTML全文浏览量: 17
- PDF下载量: 0
- 被引次数: 0
