ABSTRACT
More and more researchers are using remote sensing technology to measure real-world, on-road automobile emissions of nitric oxide (NO), one of the most important and frequently studied pollutants. Partnered with the National Institute of Water and Atmospheric Research (NIWA) in New Zealand, we aim to establish a robust NO emission factor prediction model using remote sensing data to forecast future emissions. This model can be extremely useful to local transport authorities for monitoring urban pollution levels and validating the effectiveness of existing traffic control policies. We have conducted this research using real-world data that were collected over a 10-year span between 2005 and 2015. The experimental results have shown that the vehicle emission patterns are continuously changing and the relevance of remote sensing data for future predictions decays as they get older.
We propose a 3-step machine learning approach to establish this model. Most notably we use quantile regression forest (a variation of random forest) as the base algorithm and use random forest's variable importance measure to validate and interpret the features. The model yields error rates that compare favourably to linear model based recursive partitioning and the original random forest model.
- Gary A Bishop and Donald H Stedman. 2008. A decade of on-road emissions measurements. Environmental Science & Technology 42, 5 (2008), 1651--1656.Google ScholarCross Ref
- Leo Breiman. 2001. Random forests. Machine learning 45, 1 (2001), 5--32. Google ScholarDigital Library
- David C Carslaw, Martin L Williams, James E Tate, and Sean D Beevers. 2013. The importance of high vehicle power for passenger car emissions. Atmospheric environment 68 (2013), 8--16.Google Scholar
- Irene Epifanio. 2017. Intervention in prediction measure: a new approach to assessing variable importance for random forests. BMC bioinformatics 18, 1 (2017), 230.Google Scholar
- Robin Genuer, Jean-Michel Poggi, and Christine Tuleau-Malot. 2010. Variable selection using random forests. Pattern Recognition Letters 31, 14 (2010), 2225--2236. Google ScholarDigital Library
- Liu Huan and He Kebin. 2012. Traffic Optimization: A New Way for Air Pollution Control in China's Urban Areas. Environmental Science & Technology 46, 11 (2012), 5660--5661. PMID: 22612715.Google ScholarCross Ref
- Nicolai Meinshausen. 2006. Quantile regression forests. Journal of Machine Learning Research 7, Jun (2006), 983--999. Google ScholarDigital Library
- J Metcalfe, J Bluett, and G Kuschel. 2013. Comparison of trends in predicted (VEPM) and real-world (RSD) emissions for the light duty fleet 2003 to 2011. (2013). https://www.nzta.govt.nz/assets/Highways-Information-Portal/Technical-disciplines/Air-and-climate/Planning-and-assessment/Vehicle-emissions-prediction-model/VEPM-trends-report-Dec13-Final.pdfGoogle Scholar
- Orbital. 2009. Second National In-Service Emissions Study (NISE2) Light Duty Petrol Vehicle Emissions Testing, by Roads and Traffic Authority of NSW, RTA.07.2828.0309. (2009). http://155.187.2.69/atmosphere/transport/pubs/nise2-report-09.docGoogle Scholar
- Robin Smit and Elizabeth Somervell. 2015. The use of remote sensing to enhance motor vehicle emission modelling in New Zealand. (2015). https://www.niwa.co.nz/sites/niwa.co.nz/files/AKL2015-012%20RSD%20evaluation%20%2800000003%29.pdfGoogle Scholar
- Achim Zeileis, Torsten Hothorn, and Kurt Hornik. 2008. Model-Based Recursive Partitioning. Journal of Computational and Graphical Statistics 17, 2 (2008), 492--514.Google ScholarCross Ref
- Y Zhang and GA Bishop. 1994. AUTOMOBILE EMISSIONS ARE STATISTICALLY GAMMA-DISTRIBUTED. Environmental Science & Technology 28, 7 (1994).Google Scholar
- Yi Zhang, Donald H Stedman, Gary A Bishop, Paul L Guenther, and Stuart P Beaton. 1995. Worldwide on-road vehicle exhaust emissions study by remote sensing. Environmental science & technology 29, 9 (1995), 2286--2294.Google Scholar
Index Terms
- Vehicle emission prediction using remote sensing data and machine learning techniques
Recommendations
Machine Learning and Deep Learning in Remote Sensing and Urban Application: A Systematic Review and Meta-Analysis
GEOIT4W-2020: Proceedings of the 4th Edition of International Conference on Geo-IT and Water Resources 2020, Geo-IT and Water Resources 2020Machine learning (ML) and the latest deep learning (DL) algorithms have been widely used lately in remotely sensed data analysis. Urban management has also made such developments in artificial intelligence (AI) techniques, but not with the same degree ...
Geological mapping using remote sensing data
Machine learning algorithms (MLAs) are a powerful group of data-driven inference tools that offer an automated means of recognizing patterns in high-dimensional data. Hence, there is much scope for the application of MLAs to the rapidly increasing ...
Characteristics of Remote Sensing Emission Spectra of Composite Igneous Rocks
ETTANDGRS '08: Proceedings of the 2008 International Workshop on Education Technology and Training & 2008 International Workshop on Geoscience and Remote Sensing - Volume 01Analysis of rock spectrum is a fundamental work for rocks and minerals detection by using remote sensing. Our work is focused on characteristics of composite igneous rocks within 8~14um, and studying the relationship between emissivity of rocks and ...
Comments