Optimizing Intelligent Parking Decisions using Q-Learning
DOI:
https://doi.org/10.62051/88c5p832Keywords:
Transportation; parking spaces; Q-learning; future prediction; decision support.Abstract
With urbanization progressing rapidly, the availability of transportation resources in large cities has become increasingly strained. Among these resources, parking spaces are crucial for maintaining the smooth functioning of cities and ensuring the efficient operation of vehicles. However, due to spatial limitations, it is not always feasible to continuously increase the number of parking spaces. Hence, accurate prediction of future parking space demand has become imperative. The purpose of this paper is to predict the future parking space demand by using a mathematical model, to provide scientific decision support for the transportation department. Specifically, the author chose a Q-learning model in reinforcement learning, using a dataset from the year beginning 2016 combined with an algorithm to make predictions about future data. The accuracy of the model is 81.31% and the mean square error is 2200.30. In addition, the author also combined the weather and holiday conditions to analyze the data box line. Through the analysis and modeling of urban parking data, the author will discuss the feasibility and effectiveness of the Q-learning model in future parking demand prediction. Through this study, it is concluded that the Q-learning model performs well in the prediction of future available parking Spaces.
Downloads
References
Chen Y, Hu B. Short-term traffic flow prediction based on the ARIMA model. Journal of Advanced Transportation, 2019.
Li Y, Zhang Y, Wang P. A deep learning approach for short-term traffic flow prediction. IEEE Transactions on Intelligent Transportation Systems, 2020, 22(1): 330-340.
Ismail M H, et al. Predicting vehicle parking space availability using a multilayer perceptron neural network. In IOP Conference Series: Materials Science and Engineering, 2021.
Qin X, Zhang Y, Wang P. A multi-task learning approach for short-term traffic flow prediction. IEEE Transactions on Intelligent Transportation Systems, 2021, 23(2): 1326-1337.
Wei Y, Zhang Y. A deep reinforcement learning approach for short-term traffic flow prediction. Transportation Research Part C: Emerging Technologies, 2022, 134: 103457.
Wang Y, Chen Y. A spectral clustering-based approach for short-term traffic flow prediction. Journal of Transportation Safety & Security, 2020, 12(2): 215-232.
Ye W, Kuang H, Li J, Lai X, Qu H. A parking occupancy prediction method incorporating time series decomposition and temporal pattern attention mechanism. IET Intelligent Transport Systems, 2024, 18(1): 58-71.
Yao Z, Yang J, Zhao S. A CNN-LSTM-based method for parking occupancy prediction. Sensors, 2020, 20(19): 5705.
Yu W, Wang Z, Guo S. A recurrent neural network model for predicting parking occupancy considering multi-source data. Sensors, 2021, 21(16): 5509.
Ma X, et al. Parking occupancy prediction based on LSTM with a multi-source attention mechanism. IEEE Transactions on Intelligent Transportation Systems, 2020, 22(3): 1605-1614.
QGIS. Satellite Image of KLCC in Kuala Lumpur, Malaysia. Satellite Image from QGIS of KLCC in Kuala Lumpur, Malaysia, 2022. Available at: https://qgis.org/ko/site/.
Downloads
Published
Conference Proceedings Volume
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
How to Cite
