Vehicle Trajectory Reconstruction Using Lagrange-Interpolation-Based Framework
Vehicle trajectory usually suffers from a large number of outliers and observation noises. This paper proposes a novel framework for reconstructing vehicle trajectories. The framework integrates the wavelet transform, Lagrange interpolation and Kalman filtering. The wavelet transform based on wavefo...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-01-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/14/3/1173 |
| _version_ | 1797318980779638784 |
|---|---|
| author | Jizhao Wang Yunyi Liang Jinjun Tang Zhizhou Wu |
| author_facet | Jizhao Wang Yunyi Liang Jinjun Tang Zhizhou Wu |
| author_sort | Jizhao Wang |
| collection | DOAJ |
| description | Vehicle trajectory usually suffers from a large number of outliers and observation noises. This paper proposes a novel framework for reconstructing vehicle trajectories. The framework integrates the wavelet transform, Lagrange interpolation and Kalman filtering. The wavelet transform based on waveform decomposition in the time and frequency domain is used to identify the abnormal frequency of a trajectory. Lagrange interpolation is used to estimate the value of data points after outliers are removed. This framework improves computation efficiency in data segmentation. The Kalman filter uses normal and predicted data to obtain reasonable results, and the algorithm makes an optimal estimation that has a better denoising effect. The proposed framework is compared with a baseline framework on the trajectory data in the NGSIM dataset. The experimental results showed that the proposed framework can achieve a 45.76% lower root mean square error, 26.43% higher signal-to-noise ratio and 25.58% higher Pearson correlation coefficient. |
| first_indexed | 2024-03-08T04:00:13Z |
| format | Article |
| id | doaj.art-0e82e6a99d4d4f798c40ae709238cdde |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-08T04:00:13Z |
| publishDate | 2024-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-0e82e6a99d4d4f798c40ae709238cdde2024-02-09T15:08:06ZengMDPI AGApplied Sciences2076-34172024-01-01143117310.3390/app14031173Vehicle Trajectory Reconstruction Using Lagrange-Interpolation-Based FrameworkJizhao Wang0Yunyi Liang1Jinjun Tang2Zhizhou Wu3School of Mechanical Engineering, Xinjiang University, Xinjiang 830017, ChinaDepartment of Mobility Systems Engineering, Technical University of Munich, 85748 Munich, GermanySchool of Traffic & Transportation Engineering, Central South University, Changsha 410017, ChinaSchool of Mechanical Engineering, Xinjiang University, Xinjiang 830017, ChinaVehicle trajectory usually suffers from a large number of outliers and observation noises. This paper proposes a novel framework for reconstructing vehicle trajectories. The framework integrates the wavelet transform, Lagrange interpolation and Kalman filtering. The wavelet transform based on waveform decomposition in the time and frequency domain is used to identify the abnormal frequency of a trajectory. Lagrange interpolation is used to estimate the value of data points after outliers are removed. This framework improves computation efficiency in data segmentation. The Kalman filter uses normal and predicted data to obtain reasonable results, and the algorithm makes an optimal estimation that has a better denoising effect. The proposed framework is compared with a baseline framework on the trajectory data in the NGSIM dataset. The experimental results showed that the proposed framework can achieve a 45.76% lower root mean square error, 26.43% higher signal-to-noise ratio and 25.58% higher Pearson correlation coefficient.https://www.mdpi.com/2076-3417/14/3/1173vehicle trajectory reconstructionoutlier detectionLagrange interpolationfilter denoisingNGSIM |
| spellingShingle | Jizhao Wang Yunyi Liang Jinjun Tang Zhizhou Wu Vehicle Trajectory Reconstruction Using Lagrange-Interpolation-Based Framework Applied Sciences vehicle trajectory reconstruction outlier detection Lagrange interpolation filter denoising NGSIM |
| title | Vehicle Trajectory Reconstruction Using Lagrange-Interpolation-Based Framework |
| title_full | Vehicle Trajectory Reconstruction Using Lagrange-Interpolation-Based Framework |
| title_fullStr | Vehicle Trajectory Reconstruction Using Lagrange-Interpolation-Based Framework |
| title_full_unstemmed | Vehicle Trajectory Reconstruction Using Lagrange-Interpolation-Based Framework |
| title_short | Vehicle Trajectory Reconstruction Using Lagrange-Interpolation-Based Framework |
| title_sort | vehicle trajectory reconstruction using lagrange interpolation based framework |
| topic | vehicle trajectory reconstruction outlier detection Lagrange interpolation filter denoising NGSIM |
| url | https://www.mdpi.com/2076-3417/14/3/1173 |
| work_keys_str_mv | AT jizhaowang vehicletrajectoryreconstructionusinglagrangeinterpolationbasedframework AT yunyiliang vehicletrajectoryreconstructionusinglagrangeinterpolationbasedframework AT jinjuntang vehicletrajectoryreconstructionusinglagrangeinterpolationbasedframework AT zhizhouwu vehicletrajectoryreconstructionusinglagrangeinterpolationbasedframework |