An Affordable Vehicle-Mounted Sensing Solution for Mobile Air Quality Monitoring

Santana, Pedro; Almeida, Alexandre; Mariano, Pedro; Correia, Carolina; Martins, Vânia; Almeida, Susana Marta

doi:10.23919/splitech49282.2020.9243697

Cited by 6 publications

(6 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other studies have investigated low-cost sensor performance when mounted to vehicles, but not with mobile reference or higher-grade monitors for comparison, and with differing levels of sophistication. 18,61,64,65 In this study, we tested low-cost sensor nodes containing optical PM sensors and electrochemical sensors sampling from a car while mobile, against reference and "mid-cost" instruments, also mounted to the car. Three sensor nodes were housed in a roof box on top of the car.…”

Section: ■ Introductionmentioning

confidence: 99%

Hyperlocal Air Pollution in London: Validating Low-Cost Sensors for Mobile Measurements from Vehicles

Russell,

Frederickson,

Sørensen

et al. 2024

ACS EST Air

View full text Add to dashboard Cite

This study assesses low-cost sensors (LCS) for the mobile monitoring of air quality, which has thus far been scarcely investigated. NO 2 and PM 2.5 were measured using LCS and highergrade instruments while driving across various environments in London (943 km) and northern Europe (2923 km), including urban areas, motorways, and tunnels. The data were classified according to the environment where the measurements were carried out, and the performances of LCS and higher-grade instruments were compared. Results indicate that the performances of the sensors were influenced by the rate of change in pollutant concentration in different environments and not by vehicle speed. Excluding tunnel environments, overall, the particulate matter sensors correlated better with their higher-grade instrument than the electrochemical (EC) sensors, with R 2 values from 0.90−0.96 in the different environments, compared with 0.39−0.72 for the EC sensors. Tunnels presented a unique opportunity to test the time response of the systems, given the rapid change in concentration upon entering, and all sensors showed limited response times. This is the first time that EC NO 2 sensors have been rigorously tested against reference monitors while mobile. Their absolute measurements appear unaffected by movement; however, their time resolution may not be high enough for mobile monitoring in highly variable environments.

show abstract

Section: ■ Introductionmentioning

confidence: 99%

Hyperlocal Air Pollution in London: Validating Low-Cost Sensors for Mobile Measurements from Vehicles

Russell,

Frederickson,

Sørensen

et al. 2024

ACS EST Air

View full text Add to dashboard Cite

show abstract

“…As an illustration, Santana et al. (2020) developed a cost‐effective sensing solution that can be mounted on vehicles to monitor air quality in motion. Similarly, Alessandroni et al.…”

Section: Introductionmentioning

confidence: 99%

“…In recent years, there has been a notable increase in the utilization of vehicles as mobile sensor carriers in various literature and advanced platforms. As an illustration, Santana et al (2020) developed a costeffective sensing solution that can be mounted on vehicles to monitor air quality in motion. Similarly, Alessandroni et al (2015) employed mobile devices affixed to the car cabin to detect road roughness.…”

mentioning

confidence: 99%

Revealing the mobile sensing powers of semi‐random and deterministic “drive‐by” vehicle fleets

Jiang,

Wang,

Yao

et al. 2023

Transactions in GIS

View full text Add to dashboard Cite

The utilization of mobile sensing through the deployment of sensors on third‐party vehicles that pass by an area offers notable benefits in terms of spatiotemporal coverage, precision, timeliness, and cost‐effectiveness for monitoring urban environments. However, the diverse travel behaviors exhibited by different types of vehicle fleets result in varying levels of sensing powers. Drawing inspiration from these phenomena, this study quantifies and compares the travel patterns and sensing powers of bus and taxi fleets. The experimental findings unveiled several noteworthy laws, such as a diminishing rate of increase in sensing power with the stable increasing vehicle number, the high sensing power exhibited by taxis, the comparable spatial coverage between buses, and the relatively low uncertainty associated with buses' sensing power. Furthermore, the impact of travel patterns (i.e., semi‐random behavior of taxis and deterministic behavior of buses) on the powers of mobile sensing was revealed. The aforementioned findings provide valuable insights for the development of efficient mobile sensing schemes and hold significant relevance for the implementation of smart city infrastructure.

show abstract

“…The research reported here is part of the ExpoLIS project. Our goal in the ExpoLIS project is to deploy a network of mobile LCS to monitor air quality and to develop a set of software tools [28]. This paper is an extended version of our previous work [18] aiming at developing a pollution prediction software tool to assist the population.…”

Section: Introductionmentioning

confidence: 99%

On the automated learning of air pollution prediction models from data collected by mobile sensor networks

Mariano

Almeida²,

Santana

2021

Energy Sources, Part A: Recovery, Utilization, and Environmenta

Self Cite

View full text Add to dashboard Cite

This paper addresses the problem of automated learning of air pollution predictive models that were trained using information gathered by a set of mobile low cost sensors. Concretely, fast to compute machine learning methods (Decision Trees and Support Vector Machines) were used to build regression models that predict air pollution levels for a given location. The models were trained using the data collected by the OpenSense project, in particular, number of particulate matter, particle diameter and lung deposited surface area (LDSA). We examined two different sets of attributes: one based on a geographical description of the location under analysis (e.g., distribution of households and roads), and another based on a time series of past air pollution observations in that location. Overall, we have found out that past measures lead to better pollution predictions. The best R 2 score was 0.751 obtained with the model that predicts LDSA and was trained with the data set with time series attributes, while the worst R 2 was 0.009 obtained with the geographical data set to predict number of particles. The performance of the best model is on par with similar air pollution systems. Moreover it can be used in a production system that requires frequent updates.

show abstract

An Affordable Vehicle-Mounted Sensing Solution for Mobile Air Quality Monitoring

Cited by 6 publications

References 7 publications

Hyperlocal Air Pollution in London: Validating Low-Cost Sensors for Mobile Measurements from Vehicles

Hyperlocal Air Pollution in London: Validating Low-Cost Sensors for Mobile Measurements from Vehicles

Revealing the mobile sensing powers of semi‐random and deterministic “drive‐by” vehicle fleets

On the automated learning of air pollution prediction models from data collected by mobile sensor networks

Contact Info

Product

Resources

About