The goal of IJTTE is to broaden the knowledge of traffic and transport experts and academicians by promoting free access and provide valuable insight to transport related information, research and ideas. IJTTE will serve as a great resource for researchers and students across the globe.
Volume List  / Volume 12 (2)

Article

DANGER DETECTION FOR CYCLISTS WITH MACHINE LEARNING (IN THE CITY OF COPENHAGEN)

DOI: 10.7708/ijtte2022.12(2).09


12 / 2 / 272-290 Pages

Author(s)

Mara Alena Lehmann - Hochschule Esslingen – University of Applied Sciences, Flandernstr. 101, 73732 Esslingen am Neckar, Germany -

Daniel Pascal Mair - Hochschule Esslingen – University of Applied Sciences, Flandernstr. 101, 73732 Esslingen am Neckar, Germany -

Gabriele Stefanie Gühring - Hochschule Esslingen – University of Applied Sciences, Flandernstr. 101, 73732 Esslingen am Neckar, Germany -


Abstract

This paper offers several ways to classify time series data recorded by cyclists in an urban area like Copenhagen to predict and classify dangerous situations and areas. Therefore, several neural networks used a training dataset of bicycle trips consisting of position data and associated system modes derived from a Support Vector Machine. The system modes indicate if cyclists are in dangerous situations. The model used position data and derived features like velocity, acceleration, angular deviation, and the deviation of the previous cycling behaviour in the respective trip. A gated recurrent neural network model achieved the best resulting accuracy of 83 % in a binary classification between accident and no danger. Through this, it was possible to determine if a bicycle accident happened due to the cyclist’s environment e.g., cobblestones, or due to their cycling behaviour. This way the dataset and the approved machine learning model can show municipality of cities which spots are currently posing a threat for cyclists. Furthermore, the developed algorithm can pose as a basis for a cyclist app that warns its user about dangerous driving behaviour or upcoming danger spots. All the developed algorithms can be transformed to other cities.


Download Article

Number of downloads: 311

Keywords:
accident analysis; bicycle position data; danger classification; neural network; trip segmentation;


Acknowledgements:

This study collaborates with the company Hövding Sverige AB which provided us with the data collected by the helmet Hövding 3.


References:

Cho, K.; van Merrienboer, B.; Bahdanau, D.; Bengio, Y. 2014. On the Properties of Neural Machine Translation: Encoder-Decoder Approaches. Available from Internet: https://arxiv.org/pdf/1409.1259.

 

Cycling Embassy of Denmark. 2018. The Bicycle Account 2018 Copenhagen City of Cyclists. Available from internet: https://cyclingsolutions.info/wp-content/uploads//2020/12/CPH-Bicycle-Account-2018.pdf.

 

Chollet, F. 2018. Deep Learning with Python. Manning Publications Co. USA. 361 p.

 

Dabiri, S.; Heaslip, K. 2018. Inferring transportation modes from GPS trajectories using a convolutional neural network, Transportation Research Part C: Emerging Technologies 86: 360–371.

 

Ester, M.; Kriegel, H.-P.; Sander, J.; Xu, X. 1996. A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise, KDD 96(34): 226-231.

 

Goodfellow, I.; Bengio, Y.; Courville, A. 2016. Deep Learning. MIT Press. USA. 800 p.

 

Google. 2022a. Google Maps/Google Earth (55.680494. 12.587165). Available from Internet: https://www.google.de/maps/place/55%C2%B040'49.8%22N+12%C2%B035'13.8%22E/@55.6806489,12.5869092,53m/data=!3m1!1e3!4m5!3m4!1s0x0:0xd3fe721709c52fa6!8m2!3d55.680494!4d12.587165.

 

Google. 2022b. Google Maps/Google Earth (55.686909, 12.563077). Available from Internet: https://www.google.de/maps/place/55%C2%B041'12.9%22N+12%C2%B033'47.1%22E/@55.6869197,12.5628097,53m/data=!3m1!1e3!4m13!1m7!3m6!1s0x0:0xae13b3b7bd52ac1a!2zNTXCsDQxJzEyLjkiTiAxMsKwMzMnNDcuMSJF!3b1!8m2!3d55.686909!4d12.563077!3m4!1s0x0:0xae13b3b7bd52ac1a!8m2!3d55.686909!4d12.563077.

 

Google. 2022c. Google Street View (55.680494. 12.587165). Available from Internet: https://www.google.de/maps/@55.6805485,12.5871428,3a,75y,158.7h,76.44t/data=!3m6!1e1!3m4!1sPCivWZTPWpm_D39C2vGkGA!2e0!7i16384!8i8192.

 

Google. 2022d. Google Street View (55.686909, 12.563077). Available from Internet: https://www.google.de/maps/@55.6869848,12.5630764,3a,75y,173.75h,80.98t/data=!3m6!1e1!3m4!1sTNadfLNzcXunbSEQUjRuXw!2e0!7i16384!8i8192.

 

Grus, J. 2019. Data science from Scratch. O'Reilly Media, Inc. USA. 376 p.

 

Hochreiter, S.; Schmidhuber, J. 1997. Long short-term memory, Neural Computation 9(8): 1735–1780.

 

Holmgren, J.; Knapen, L.; Olsson, V.; Masud, A. P. 2020. On the use of clustering analysis for identification of unsafe places in an urban traffic network, Procedia Computer Science 170: 187-194.

 

International Transport Forum. 2013. Cycling, Health and Safety. Organisation for Economic Co-operation and Development Publishing. France. 248 p.

 

Jiang, X.; Souza, E. N. d.; Pesaranghader, A.; Hu, B.; Silver, D. L.; Matwin, S. 2017. TrajectoryNet: An Embedded GPS Trajectory Representation for Point-based Classification Using Recurrent Neural Networks. Available from Internet: http://arxiv.org/pdf/1705.02636v2.

 

Karakaya, A.S.; Hasenburg, J.; Bermbach, D. 2020. SimRa: Using crowdsourcing to identify near miss hotspots in bicycle traffic, Pervasive and Mobile Computing 67: 101-197.

 

Kingma, D. P.; Ba, J. 2014. Adam: A Method for Stochastic Optimization. Available from Internet: https://arxiv.org/pdf/1412.6980.

 

Krizhevsky, A.; Sutskever, I.; Hinton, G. E. 2017. Imagenet classification with deep convolutional neural networks, Communications of the ACM 60(6): 84–90.

 

Lindqvist, S.; Roos, J. 2020. Identification of areas with increased risk of accidents for cyclists based on bicycle helmet data [In Norwegian: Identifiering av områden med förhöjd olycksrisk för cyklister baserad på cykelhjälmsdata]. Available from Internet: https://www.diva-portal.org/smash/get/diva2:1480296/FULLTEXT01.pdf.

 

MacQueen, J. 1967. Some methods for classification and analysis of multivariate observations. In Proceedings of the fifth Berkeley symposium on mathematical statistics and probability, 281–297.

 

Nisbet, A. 2020. Open Topo Data. Available from Internet: https://www.opentopodata.org/#public-api.

 

Saiprasert, C.; Pholprasit, T.; Thajchayapong, S. 2017. Detection of Driving Events using Sensory Data on Smartphone, International Journal of Intelligent Transportation Systems Research 15(1): 17–28.

 

Sinnott, R. W. 1984. Virtues of the Haversine, Sky and Telescope 68(2): 158.

 

Steinwart, I.; Christmann, A. 2008. Support Vector Machines. Springer. USA. 601 p.

 

STRADA. 2021. About the Strada accident database [In Swedish: Om olycksdatabasen Strada]. Available from Internet: https://www.transportstyrelsen.se/STRADA.

 

Sun, S.; Chen, J.; Sun, J. 2019. Traffic congestion prediction based on GPS trajectory data, International Journal of Distributed Sensor Networks 15(5): 155014771984744.

 

TensorFlow. 2021a. Time series forecasting; TensorFlow Core. Available from Internet: https://www.tensorflow.org/tutorials/structured_data/time_series?hl=en.

 

TensorFlow. 2021b. Module: tf.keras; TensorFlow Core v2.7.0. Available from Internet: https://www.tensorflow.org/api_docs/python/tf/keras.

 

Wang, Y.; Qin, K.; Chen, Y.; Zhao, P. 2018. Detecting Anomalous Trajectories and Behavior Patterns Using Hierarchical Clustering from Taxi GPS Data, ISPRS International Journal of Geo-Information 7(1): 25.

 

Waskom, M. L. 2021. Seaborn: statistical data visualization, Journal of Open Source Software 6(60): 3021.

 

Watson, A.; Watson, B.; Vallmuur, K. 2015. Estimating under-reporting of road crash injuries to police using multiple linked data collections, Accident; Analysis and Prevention 83: 18–25.


Quoted IJTTE Works

There is no quoted studies.



Related Keywords

PREDICTION OF EMISSIONS FROM BIODIESEL FUELED TRANSIT BUSES USING ARTIFICIAL NEURAL NETWORKS