A machine learning approach for traffic-noise annoyance assessment

Bravo-Moncayo, Luis; Lucio-Naranjo, José; Chávez, M.; Pavón-García, Ignacio; Garzón, Christiam

doi:10.1016/j.apacoust.2019.07.010

Cited by 48 publications

(23 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Their predictive powers have been compared through several studies, both in laboratory conditions and from social survey data, and extensive description and discussion can be found in the literature [27][28][29][30]. Annoyance models for single or combined sources have typically focused on different disturbing source types, such as: road traffic [28,29,[31][32][33][34][35][36], railways [37,38], tramways [35,39], industrial noise [27][28][29]31,40,41], and aircraft noise [36,42]. Consequently, little is said about "wanted" sounds or sounds of preference.…”

Section: Introductionmentioning

confidence: 99%

Acoustic Design Criteria in Naturally Ventilated Residential Buildings: New Research Perspectives by Applying the Indoor Soundscape Approach

et al. 2019

View full text Add to dashboard Cite

The indoor-outdoor connection provided by ventilation openings has been so far a limiting factor in the use of natural ventilation (NV), due to the apparent conflict between ventilation needs and the intrusion of external noise. This limiting factor impedes naturally ventilated buildings meeting the acoustic criteria set by standards and rating protocols, which are reviewed in this paper for residential buildings. The criteria reflect a general effort to minimize noise annoyance by reducing indoor sound levels, typically without a distinction based on a ventilation strategy. Research has developed a number of solutions, discussed here, that try to guarantee ventilation without compromising façade noise insulation, but, currently, none have been adopted on a large scale. This concept paper highlights the main limits of the current approach. First, a fragmented view towards indoor environmental quality has not included consideration of the following acoustic criteria: (i) how buildings are designed and operated to meet multiple needs other than acoustical ones (e.g., ventilation, visual, and cooling needs) and (ii) how people respond to multiple simultaneous environmental factors. Secondly, the lack of a perceptual perspective has led acoustic criteria to neglect the multiple cognitive and behavioral factors impinging on comfort in naturally ventilated houses. Indeed, factors such as the connection with the outside and the sense of control over one’s environment may induce “adaptive acoustic comfort” opportunities that are worth investigating. The mere use of different sound level limits would not be enough to define criteria tailored to the complex user–building interaction that occurs under NV conditions. More holistic and human-centered approaches are required to guarantee not only neutrally but even positively perceived indoor acoustic environments. For this reason, this paper considers this apparent conflict from a soundscape viewpoint, in order to expose still unexplored lines of research. By underpinning a perceptual perspective and by contextualizing it, the indoor soundscape approach provides a framework capable of overcoming the limits of the traditional noise control approach. This could provide the opportunity to foster a wider adoption of NV as a passive design strategy that enhances user health and well-being, while enabling low-cost, and low-energy cooling and ventilation, thereby contributing to current climate change challenges.

show abstract

Section: Introductionmentioning

confidence: 99%

Acoustic Design Criteria in Naturally Ventilated Residential Buildings: New Research Perspectives by Applying the Indoor Soundscape Approach

et al. 2019

View full text Add to dashboard Cite

show abstract

“…As principais funcionalidades dos métodos nos últimos 5 anos foram a classificação de tipos de ruídos [Kumari et al 2019], a predição de indicadores de níveis de ruído [Torija and Ruiz 2015], a classificação de níveis de ruído [Torija and Ruiz 2016], avaliação do incômodo causado por ruídos [Bravo-Moncayo et al 2019] e a detecção de eventos anômalos [Alías et al 2018]. Para essas aplicações, as técnicas mais usadas foram: máquinas de vetor de suporte (SVM) [Torija and Ruiz 2016] (12), redes neurais artificiais (ANN) [Torija and Ruiz 2016] (11), redes neurais convolucionais (CNN) [Kumari et al 2019] (11), florestas aleatórias [Liu et al 2020] (6), árvores de decisão [Ali Khalil et al 2019] (5), regressão [Bravo-Moncayo et al 2019] (3), regressão de vetores de suporte (SVR) [Giannakopoulos et al 2019] (3), processos gaussianos [Torija and Ruiz 2015] (2), modelos de conjunto [Nourani et al 2020] (2) e redes neurais recorrentes LSTM [Navarro et al 2020] (2).…”

Section: Aprendizado De Máquina Aplicado à Poluição Sonoraunclassified

Desafios e oportunidades da aplicação de Sistemas Ciberfísicos no monitoramento da poluição urbana

Santos¹,

Freitas²,

Teixeira³

et al. 2020

Anais Do Workshop De Computação Urbana (CoUrb 2020)

View full text Add to dashboard Cite

A ação do homem tem provocado mudanças no meio ambiente, sendo a poluição urbana uma das consequências negativas aplicadas nesse ecossistema. Com o desenvolvimento tecnológico, novas perspectivas computacionais afloram para o monitoramento ambiental. Este artigo apresenta pontos chaves da fenomenologia ambiental que podem se beneficiar da evolução promovida pela computação aplicada nos estudos da poluição, assim como realiza um levantamento de pesquisas e tecnologias utilizadas nesse contexto.

show abstract

“…It was found that the nonlinear ensemble techniques produced the best result, which improved the performance of single models with a great robustness. Apart from traffic noise level prediction, deep learning has also been applied to other traffic-related studies, such as traffic flow state estimation [24], traffic light detection and classification [25], traffic condition forecasting [26], autonomous driving [27] and traffic noise annoyance assessment, related to physical characteristics of sound and subjective perception of the person [28], etc. In the work of [29], a classification model was developed to distinguish road traffic noise and anomalous noise events.…”

Section: Introductionmentioning

confidence: 99%

Traffic Noise Prediction Applying Multivariate Bi-Directional Recurrent Neural Network

2021

View full text Add to dashboard Cite

With the drastically increasing traffic in the last decades, crucial environmental problems have been caused, such as greenhouse gas emission and traffic noise pollution. These problems have adversely affected our life quality and health conditions. In this paper, modelling of traffic noise employing deep learning is investigated. The goal is to identify the best machine-learning model for predicting traffic noise from real-life traffic data with multivariate traffic features as input. An extensive study on recurrent neural network (RNN) is performed in this work for modelling time series traffic data, which was collected through an experimental campaign at an inner city roundabout, including both video traffic data and audio data. The preprocessing of the data, namely how to generate the appropriate input and output for deep learning model, is detailed in this paper. A selection of different architectures of RNN, such as many-to-one, many-to-many, encoder–decoder architectures, was investigated. Moreover, gated recurrent unit (GRU) and long short-term memory (LSTM) were further discussed. The results revealed that a multivariate bi-directional GRU model with many-to-many architecture achieved the best performance with both high accuracy and computation efficiency. The trained model could be promising for a future smart city concept; with the proposed model, real-time traffic noise predictions can be potentially feasible using only traffic data collected by different sensors in the city, thanks to the generated big data by smart cities. The forecast of excessive noise exposure can help the regulation and policy makers to make early decisions, in order to mitigate the noise level.

show abstract

A machine learning approach for traffic-noise annoyance assessment

Cited by 48 publications

References 23 publications

Acoustic Design Criteria in Naturally Ventilated Residential Buildings: New Research Perspectives by Applying the Indoor Soundscape Approach

Acoustic Design Criteria in Naturally Ventilated Residential Buildings: New Research Perspectives by Applying the Indoor Soundscape Approach

Desafios e oportunidades da aplicação de Sistemas Ciberfísicos no monitoramento da poluição urbana

Traffic Noise Prediction Applying Multivariate Bi-Directional Recurrent Neural Network

Contact Info

Product

Resources

About