Academic Journal of Engineering and Technology Science, 2024, 7(4); doi: 10.25236/AJETS.2024.070418.
Zhengen Lv1, Qi Liu2, Yicheng Shen3, Pengyang Wei4, Zhicheng Pan5
1College of Mechanical and Control Engineering, Guilin University of Technology, Guilin, China
2School of Energy and Electricity Engineering, Qinghai University, Sining, China
3College of Mechanical and Control Engineering, Guilin University of Technology, Guilin, China
4School of Biomedical Engineering and Technology, Tianjin Medical University, Tianjin, China
5School of Petroleum and Natural Gas Engineering, Changzhou University, Changzhou, China
With the acceleration of urbanisation, the problem of traffic congestion is becoming more and more prominent, and effective traffic management and optimisation is increasingly becoming an important issue in urban operation. In this paper, we will focus on the relevant data and discuss in depth how to use vehicle trajectories to estimate the operation cycle of traffic signals. In this paper, the K-means model is first used to analyse the periodicity of five specified intersections in a particular direction based on the vehicle trajectories within one hour and accurately calculate the periodicity of traffic signals at these intersections in that direction. The clustering model is continuously adjusted through iterative optimisation to minimise intra-class distances and maximise inter-class distances. Then, the peak detection method is used to solve the red light phase duration and calculate the continuous stop state interval to estimate the green light duration. In this paper, for dynamic cycle change detection, peak analysis is used to identify the major stop duration peaks and the CUSUM method is further applied to detect cycle change points. This helps in mapping the vehicle trajectory using the provided data and confirming the approximate orientation of the vehicle as a means of categorising the data and constructing an accurate cycle model.
Cluster analysis, Dynamic cycle detection, Data visualisation, CUSUM, Peak analysis
Zhengen Lv, Qi Liu, Yicheng Shen, Pengyang Wei, Zhicheng Pan. A study on the problem of predicting traffic signal cycles by travelling trajectories. Academic Journal of Engineering and Technology Science (2024) Vol. 7, Issue 4: 122-128. https://doi.org/10.25236/AJETS.2024.070418.
[1] Q. Z. Wu, J. Zhou. Research on the signal cycle of traffic signals in multi-intersection road sections based on fuzzy control[J]. Industrial Control Computer, 2019, Vol. 32(6): 81-82, 85
[2] Y. T. Luo, T. Wang, M.N. Yang, Y.K. Zhang. A visual analysis method for vehicle behaviour based on historical driving trajectory set [J]. Computer Science, 2021, Vol. 48 (9): 86-94
[3] R. Lv. Research on trajectory and driving behaviour of transport vehicles based on clustering [D]. Chang’an University, 2022
[4] Wu, Yi; Wang, Wei; Wang, Xuejun. Convergence of the CUSUM estimation for a mean shift in linear processes with random coefficients [J]. Computational Statistics, 2024: 1-26
[5] Jevgenijs Ivanovs; Kazutoshi Yamazaki. A series expansion formula of the scale matrix with applications in CUSUM analysis [J]. Stochastic Processes and their Applications, 2024, Vol.170: 104300
[6] Ahad, Nauman; Davenport, Mark A.; Xie, Yao. Data-adaptive symmetric CUSUM for sequential change detection [J]. Sequential Analysis-Design Methods and Applications, 2024, Vol. 43(1): 1-27