Improving parking availability information using deep learning techniques

Urban traffic currently affects the quality of life in cities and metropolitan areas as the problem becomes ever more aggravated by parking issues: congestion increases due to individuals looking for slotsto park their vehicles. An Internet of Things approach allows drivers to know the state of the...

Descripción completa

Detalles Bibliográficos
Autores: Arjona Martínez, Jamie|||0000-0003-2585-2392, Linares Herreros, María Paz|||0000-0002-6756-8931, Casanovas Garcia, Josep|||0000-0002-7026-0962, Vázquez Giménez, Juan José
Tipo de recurso: artículo
Fecha de publicación:2020
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/343786
Acceso en línea:https://hdl.handle.net/2117/343786
https://dx.doi.org/10.1016/j.trpro.2020.03.113
Access Level:acceso abierto
Palabra clave:Urban transportation
Deep learning
Electronic villages (Computer networks))
Automobile parking
parking availability forecast
deep learning
smart cities
Transport urbà
Aprenentatge profund
Ciutats digitals (Xarxes d'ordinadors)
Automòbils -- Aparcament
Àrees temàtiques de la UPC::Enginyeria civil::Infraestructures i modelització dels transports
Descripción
Sumario:Urban traffic currently affects the quality of life in cities and metropolitan areas as the problem becomes ever more aggravated by parking issues: congestion increases due to individuals looking for slotsto park their vehicles. An Internet of Things approach allows drivers to know the state of the parkingsystem in real time through wireless networks of sensor devices. This work focuses on studying the data generated by parking systems in order to develop predictive models that generate forecasted information. This can be useful in improving the managementof parking areas, especially on-streetparking, while having an important effect on urban traffic. This research begins by looking at thestate of the art in predictive methods based on machine learning for time series. Similar studies and proposed solutions for parking predictionare described in terms of the technology and current state-of-the-art predictive models. This paper then introduces the recurrent neural network methodsthat were usedin this research,namely Long Short-Term Memory and Gated Recurrent Unit, as well as the models developed according to real scenarios in different cities. In order to improve the quality of the models, exogenous variables like hourly weather and calendar effects are taken into account,and the baseline models are compared to the models that usedthis information. Finally, the preliminary encouraging results are described, followed by suggestions for corresponding future work