Specialists from the University of Glasgow believe self-driving cars “need to learn the language of cyclists”, with their research suggesting such improvements are necessary to help autonomous vehicles safely share the roads with those riding bicycles.

In a paper titled ‘Keep it Real: Investigating Driver-Cyclist Interaction in Real-World Traffic’, which will be published later in 2023 and was today reported by The Herald newspaper, researchers looked to unpick the relationship between cyclists and automated vehicles, saying there had been “comparatively little” research into how self-driving technology can keep cyclists safe.

Professor Stephen Brewster of the university’s School of Computing Science said there had been “a lot of research in recent years on building safety features into autonomous vehicles to help keep pedestrians safe”, something that needs to be repeated with cyclists.

> “These are completely safe autonomous vehicles”: Cyclist spots driverless car using cycle lane

“Cars and bikes share the same spaces on the roads, which can be dangerous – between 2015 and 2020, 84 per cent of fatal bike accidents involved a motor vehicle, and there were more than 11,000 collisions,” he said.

“There has been a lot of research in recent years on building safety features into autonomous vehicles to help keep pedestrians safe, but comparatively little on how automated vehicles can safely share the road with cyclists.

“That’s a cause for concern as automated vehicles become more commonplace on the roads. While pedestrians tend to meet automated vehicles in highly controlled situations like road crossings, cyclists ride alongside cars for prolonged periods and rely on two-way interactions with drivers to determine each other’s intentions.

“It’s a much more complicated set of behaviours, which makes it a big challenge for future generations of automated vehicles to tackle. Currently, self-driving cars offer very little direct feedback to cyclists to help them make critically important decisions like whether it’s safe to overtake or to switch lanes. Adding any guesswork to the delicate negotiations between car and bike has the potential to make the roads less safe.”

> Tesla car in Full Self-Driving Beta almost rams cyclist

Brewster’s team studied the ways drivers and cyclists directly and indirectly communicate in real-world situations. From the research they have formed recommendations for future generations of automated vehicles.

The researchers suggest the vehicles’ intentions could be displayed on their exteriors, for example displaying animations signalling intention to speed up, slow down, give way or manoeuvre.

At the other end of the relationship they suggest cyclists could wear ‘smart glasses’ communicating the vehicle’s intentions to them, for example when coloured LEDs on the car light up to signal right of way is up for negotiation a vibration could be sent to the glasses as a non-verbal message.

> Tesla using Full Self-Driving Beta crashes into cycle lane bollard…weeks after Elon Musk’s zero collisions claim

The paper’s co-author, Ammar Al-Taie, said he hopes the research will inform autonomous vehicle designers, encouraging them to develop “new ways that self-driving cars can work safely alongside cyclists by speaking their language”.

“Just like spoken languages, communication between cyclists and drivers varies from country to country. We’re very conscious that this paper focuses specifically on UK roads – any future developments will need to take into account the differences in drivers’ and cyclists’ interactions across the world.”

The research will be presented, at the ACM Conference on Human factors in Computing Systems, in Germany next week.