School of Transportation Management,People's Public Security University of China,Beijing 100038,China
Show less
文章历史+
Received
Accepted
Published
2024-04-07
Issue Date
2026-06-15
PDF (2781K)
摘要
单全球导航卫星系统(GNSS)环境下车辆速度检测存在显著噪声干扰,且数据采样的稀疏性会加剧速度计算的不稳定性。为此,将单GNSS下车辆速度估计构建为基于时空相关性的速度期望优化问题,并提出一种自监督双向长短时记忆(LSTM)算法求解。首先,该算法构建稀疏数据时空特征提取LSTM网络,引入时间门和空间门控函数来分析稀疏、不等间距的车辆检测数据中速度时空关联的变化,提取速度时空特征嵌入向量;其次,噪声双向抑制的车辆速度估计LSTM网络分别从前向、后向来分析车辆速度变化趋势,准确实现噪声清除与速度估计;最后,以GPS信号为例对算法的车辆速度估计性能进行了实验验证,结果表明:提出的算法在采样间隔为1~10 min的稀疏数据下,去除速度数据中的噪声平均值为26.73 dB PSNR,与LWR、EnKF、Noise2Void算法相比平均高28.93%,估计速度的准确度平均高2.02%。
Abstract
The vehicle speed detection in the single Global Navigation Satellite System (GNSS) environment is subject to significant noise interference, and the sparsity of data sampling will further exacerbate the instability of speed calculation. To address this issue, the vehicle speed estimation under the single GNSS scenario is formulated as a speed expectation optimization problem based on spatiotemporal correlation, and a self-supervised bidirectional Long Short-Term Memory (LSTM) algorithm is proposed for its solution. Firstly, a sparse data spatiotemporal feature extraction LSTM is constructed by this algorithm, where time gate and spatial gate functions are introduced to analyze the changes in speed spatiotemporal correlations in sparse and unequal-interval vehicle detection data, and embedded vectors for speed spatiotemporal features are extracted. Secondly, the trend of vehicle speed changes is analyzed from both forward and backward directions by the noise bidirectional suppression LSTM network for vehicle speed estimation, enabling the accurate achievement of noise elimination and speed estimation. Finally, experimental verification of the vehicle speed estimation performance of the proposed algorithm was conducted using GPS signals as an example. The results show that an average noise reduction of 26.73 dB PSNR was achieved by the proposed algorithm in sparse speed data with sampling intervals ranging from 1 minute to 10 minutes, which is 28.93% higher than that achieved by the LWR, EnKF, and Noise2Void algorithms on average. Additionally, the speed estimation accuracy of the proposed algorithm is 2.02% higher on average.
在去除车辆速度数据中的噪声误差方面,对同一类型噪声,SSB-LSTM算法在不同噪声水平下的PSNR取平均值,算法对高斯噪声、泊松噪声、尖峰噪声的平均去除量分别为21.57、22.67、18.52 dB PSNR。可以得出,SSB-LSTM去除泊松噪声的能力最强,去除尖峰噪声的能力最弱(较泊松噪声低18.31%)。随噪声水平从5 km2/h2提高至20 km2/h2,算法去除泊松噪声的PSNR增加5.39%,而去除高斯噪声和尖峰噪声的能力则没有明显的下降或上升趋势。算法对由建筑物遮挡、天气等环境干扰因素导致的车辆速度检测误差去除能力更强,对由车辆故障、急刹车等突变因素导致的速度误差去除能力略弱。
MinHai-gen, FangYu-kun, WuXia, et al. Position prediction based on empirical mode decomposition and long short-term memory under global navigation satellite system outages[J]. China Journal of Highway and Transport, 2021, 34(7): 128-139.
[3]
ShenC, ZhangY, TangJ, et al. Dual-optimization for a MEMS-INS/GPS system during GPS outages based on the cubature Kalman filter and neural networks[J]. Mechnical Systems and Signal Processing, 2019, 133:No. 106222.
[4]
YaoY Q, XuX S, ZhuC C, et al. A hybrid fusion algorithm for GPS/INS integration during GPS outages[J]. Measurement, 2017, 103: 42-51.
[5]
LehtinenJ, MunkbergJ, HasselgrenJ, et al. Noise2Noise: learning image restoration without clean data[C]∥International Conference on Machine Learning, Stockholm, Sweden, 2018: 2965-2974.
[6]
MoranN, DanS, YuZ, et al. Noisier2Noise: learning to denoise from unpaired noisy data[C]∥IEEE Conference on Computer Vision and Pattern Recognition, Seattle, USA,2020: 12061-12069.
[7]
KrullA, BuchholzT O, JugF. Noise2Void-Learning denoising from single noisy images[C]∥IEEE Conference on Computer Vision and Pattern Recognition, Long Beach, USA, 2019: 2124-2132.
LiuXue-mei, Sheng-nanChengpeng, LiHai-rui, et al. Research on text entity recognition for water project inspection based on word-character vector BiLSTM-CRF[J]. Journal of North China University of Water Resources and Electric Power, 2023(3): 9-17.
SuZhao-pin, ZhangLing, ZhangGuo-fu, et al. A speech steganalysis algorithm based on multi-feature fusion and BiLSTM[J]. Acta Electronica Sinica, 2023, 51(5): 1300-1309.
FuXiang, XiaoShuai, XuChao. Parallel compound braking strategy of vehicle driven by wheel motor[J]. Journal of Jiangsu University(Natural Science Edition), 2025, 46(1): 9-17.
[16]
GuoS, YanZ F, ZhangK, et al. Toward convolutional blind denoising of real photographs[C]∥IEEE Conference on Computer Vision and Pattern Recognition, Long Beach, USA, 2019: 1712-1722.
[17]
HeK M, ChenX L, XieS N, et al. Masked autoencoders are scalable vision learners[C]∥IEEE Conference on Computer Vision and Pattern Recognition, New Orleans,USA,2022: 15979-15988.
[18]
CuiZ Y, KeR M, PuZ Y, et al. Stacked bidirectional and unidirectional LSTM recurrent neural network for forecasting network-wide traffic state with missing values[J]. Transportation Research Part C: Emerging Technologies, 2020, 118(9): No.102674.
[19]
LvY S, DuanY J, KangW W, et al. Traffic flow prediction with big data: a deep learning approach[J]. IEEE Transactions Intelligent Transportation Systems, 2015, 16(2): 865-873.
[20]
DoL N N, VuH L, VoB Q, et al. An effective spatial-temporal attention based neural network for traffic flow prediction[J]. Transportation Research Part C: Emerging Technologies, 2019(108): 12-28.
JiaXian-guang, FengChao-qin, SuZhi-wen, et al. Forecasting for urban traffic grid clusters based on Bi-LSTM[J]. Jounal of Chongqing University, 2023, 46(9):130-141.
[23]
GravesA, JaitlyN, MohamedA R. Hybrid speech recognition with deep bidirectional lstm[C]∥IEEE Workshop on Automatic Speech Recognition and Understanding, Olomouc, Czech Republic, 2013: 273-278.