RoSPA - See.Sense Cycle Smart Brum

In 2020, 141 cyclists were killed in Great Britain, whilst 4,215 were reported to have been seriously injured. In order to increase cycling safety and work towards achieving Vision Zero - that is, eliminating all traffic fatalities and serious injuries on the roads - we need data. Specifically, we need more accurate and more comprehensive data. 

Through our pioneering research with the Royal Society for the Prevention of Accidents (RoSPA), we have shown how See.Sense data can be used as a ‘lead indicator’ to improve cyclist safety. Funded by the Department for Transport, the study actively involved over 200 cyclists equipped with sensor-enabled See.Sense bike lights. Over a six-month period, these participants covered over 26,000 miles and relayed billions of lines of data relating to their riding environment and activity via the light sensors. 

This comprehensive data set enabled researchers to form an extraordinarily accurate picture of locations in the city where riders most frequently experienced ‘near miss’ brake or swerve incidents that did not result in accidents and therefore were usually not reported to police.

See.Sense lights contain patented, AI-enabled sensor technology that monitors the rider's environment up to 800 times per second. This provides highly granular anonymised insights into the rider's experience, including braking, swerving, collisions, and road surface conditions. Reports can also be manually submitted through the See.Sense app, enabling users to pinpoint and explain any problems experienced. 

By utilising the crowdsourced data collected from the See.Sense lights, the study showed how it was possible to identify the most hazardous cycling areas, that can then be prioritised for interventions to proactively address cycling safety. 

Project Results: 

• Participants cycled over 26,000 miles, representing 7.9 Billion individual sensor readings for swerving, braking and surface condition.
• The comprehensive data set enabled researchers to form an extraordinarily accurate picture of locations in Birmingham where riders most frequently experienced ‘near miss’ brake or swerve incidents that did not result in collisions, and therefore were usually not reported to police.

• By cross-referencing this data with historic ‘STATS19’ police reports outlining collision locations in the Birmingham area, RoSPA and See.Sense were able to build a predictive model showing where cycle collisions are more likely to occur.  
• See.Sense and RoSPA provided compelling evidence that cyclists are 2.4 times more likely to experience a heavy brake or swerve event in the immediate vicinity of officially recorded collision locations.
• This project was featured in numerous publications such as Cycling Industry News and is a Winner of the Rees Jeffreys Road Fund Competition (Stage 1). 

“We would propose that this could form the basis for a useful investigative tool to quickly identify the most hazardous cycling areas. Or alternatively, this could be used as a tool to analyse an area based on other indicating data, such as reports from cyclists” - RoSPA.

The findings of the project have been published by RoSPA, which can be accessed here: Cycle Smart Brum Project: Summary Report. 

David Walker, RoSPA’s Head of Road Safety, stated: 

“Cycling collisions are typically under-reported and therefore it’s vital that we understand more about their causes, so that road safety can be improved. Up until now, we’ve had to rely on ‘lag’ indicators such as the STATS19 report filed by police, usually when there has been a serious injury or death.  This is why we are really excited about this research, which highlights how the swerving and braking data forms ‘lead-indicators’ that can help cities prioritise their safety interventions, or as a tool to analyse an area based on other indicating data, such as reports from cyclists.”

This project was featured at the Cycling and Walking Innovation Conference 2021, video replay available here:

This project was recognised and platformed by the United Nations AI For Good Program. 

To find out how See.Sense technology and data services can help with your vision zero and road safety projects please email team@seesence.cc and a member of our team will be in touch.