Project Velograph: Working to make our roads safer



About the Project

Project Velograph is a joint project between the University of Adelaide's Centre for Automotive Safety Research (CASR), Bicycle Network, and Insight via Artificial Intelligence (IVAI) with the goal of making Australian roads safer for cyclists.


Over the last decade, cyclist fatalities and serious injuries have increased steadily, indicating that cyclist safety needs attention. Close passing presents one of the most dangerous risks for a cyclist while riding on the road and deters potential cyclists from taking up riding. Information about the interactions between cyclists and vehicles across the road network is scarce. Because road authorities do not know enough about the risky locations for cyclists, they cannot effectively apply safety interventions.


We are producing a comprehensive dataset of cyclist-vehicle passing distances with details about the passing distance and the location, along with a set of data processing and visualisation tools. The data is collected in a naturalistic way by volunteer cyclists to provide “real world” evidence of locations where cycling is riskier. Our software is designed to enable road authorities to view cycling risk across their network and prioritise spending on road safety interventions.


Custom non-intrusive hardware is attached to a bike, consisting of a GPS tracker and two short-range radars, which we use to pinpoint the locations where passing cars keep an unsafe distance. Volunteers upload data from their devices to the Project Velograph website, where it is processed and displayed. Each volunteer can see the raw data from their rides, while the anonymised data from all users is accumulated and used to determine dangerous road areas.


The volunteer dashboard view with a map showing their trips and passing distances

IVAI's Role

Our role in the project is to design, build and maintain the web portal and backend systems and conduct all of the data analysis. The work includes the website, user, and administration dashboards where passing distances can be uploaded, viewed, modified, and deleted. We utilise the Google Maps API to not only display information over a map but also to correct errors in GPS data uploaded to the system. We use techniques from robust and spatial statistics to identify genuine passing events and highlight problematic areas.


Impact

Cycling as a mode of transport has proven benefits to traffic congestion, cardiovascular health, the liveability of our cities, and the economy (de Hartog et al., 2010; Oja et al., 2011; National Heart Foundation of Australia, 2019). However, the number of cyclist fatalities and serious injuries (FSI) on our public roadways has increased during the last decade. Indeed, data from the Road Trauma Australia 2019 Statistical Summary (BITRE, 2020) shows that cyclist fatalities have remained fairly consistent at an average of 37 per year, while hospitalisations have been steadily increasing from 5,239 in 2010 to 7,077 in 2017 (most recent data available). While some of these hospitalisations will be the result of single-cyclist crashes, higher severity injuries are more likely to be the result of cyclist-vehicle collisions (Beck et al., 2017). Furthermore, crashes resulting from a vehicle passing a cyclist too closely are more likely to result in severe injuries compared to other types of crashes between cyclists and vehicles (Stone & Broughton, 2003; Raslavičius et al., 2017). The perception that cycling on the road is dangerous is also acting as a major deterrent to the uptake of cycling (Munro, 2017; Pooley et al., 2013).


To reverse the current trend and improve cyclist safety, road authorities need an evidence base they can use to inform and prioritise their resource allocation decisions as well as evaluate the effectiveness of the interventions they deploy. This project will deliver a platform capable of providing the necessary evidence for infrastructure planning and investment targeted at improving safety for cyclists as vulnerable road users.


References

Bureau of Infrastructure, Transport and Regional Economics (BITRE), 2020. Road Trauma Australia 2019 Statistical Summary. BITRE, Canberra.


Beck, B., Ekegren, C.L., Cameron, P., Edwards, E.R., Bucknill, A., Judson, R., Page, R., Hau, R., Stevenson, M., Gabbe, B.J., 2017. Predictors of recovery in cyclists hospitalised for orthopaedic trauma following an on-road crash. Accident Analysis & Prevention 106, 341–347. https://doi.org/10.1016/j.aap.2017.06.019


de Hartog, J.J., Boogaard, H., Nijland, H., Hoek, G., 2010. Do the Health Benefits of Cycling Outweigh the Risks? Environmental Health Perspectives 118, 1109–1116. https://doi.org/10.1289/ehp.0901747


Oja, P., Titze, S., Bauman, A., de Geus, B., Krenn, P., Reger-Nash, B., Kohlberger, T., 2011. Health benefits of cycling: a systematic review. Scandinavian Journal of Medicine & Science in Sports 21, 496–509. https://doi.org/10.1111/j.1600-0838.2011.01299.x


Pooley, C.G., Horton, D., Scheldeman, G., Mullen, C., Jones, T., Tight, M., Jopson, A., Chisholm, A., 2013. Policies for promoting walking and cycling in England: A view from the street. Transport Policy 27, 66–72. https://doi.org/10.1016/j.tranpol.2013.01.003


Munro, C., 2017. National Cycling Participation Survey 2017: Australian Capital Territory (No. AP-C91- 17). Austroads, Sydney.


Raslavičius, L., Bazaras, L., Keršys, A., Lukoševičius, V., Makaras, R., Eidukynas, V., 2017. Assessment of bicycle–car accidents under four different types of collision. Proc Inst Mech Eng H 231, 222–234. https://doi.org/10.1177/0954411917690417


Stone, M., Broughton, J., 2003. Getting off your bike: cycling accidents in Great Britain in 1990–1999. Accident Analysis & Prevention 35, 549–556. https://doi.org/10.1016/S0001-4575(02)00032-5