As DG Cities publishes analysis from the Project Endeavour self-driving car trials, Head of Research and Service Design, Ed Houghton explores some of the key findings. With safety a key concern, he looks at the role of public engagement in the wider adoption of the technology.
One of the Endeavour fleet of cars on the road in Greenwich, London
Self-driving or autonomous cars are often cited as the next major breakthrough in mobility innovation. As a result, there is increasing political and public interest in trials of the technology and exploring the potential uses of autonomous vehicles (AVs) in our towns and cities. One of the most significant areas of interest is public attitudes – people’s perception of AVs, their hopes and fears. It’s well known that the biggest barrier to adoption and acceptance of any new technology based on AI is low trust and limited knowledge. This is why, as part of the Project Endeavour consortium, a UK government funded initiative to trial autonomous vehicles, we investigated how members of the public feel about self-driving vehicles when they get the chance to experience one for themselves.
Experiencing autonomy first-hand
Through a live trial, we explored what the public thinks of self-driving cars, and crucially, if a journey in one can grow their knowledge and even help them to trust the technology more. In 2021, Project Endeavour ran a major trial in London, which was open to the public and widely publicised. We invited people to ride in the vehicle and experience a journey on the streets of a busy city neighbourhood. Participants were able to sit in a vehicle as it operated in autonomy, while supervised by two trained personnel: a Safety Driver and an Automated Control System Operator (ACSO). Both helped to ensure the ride was safe, and also answered any questions that members of the public might have about the technology and how the car was making decisions.
We ran before and after surveys online to explore how perceptions changed over time, and conducted face-to-face interviews, along with a series of online focus groups with the public. Alongside the trial, we ran a national study of attitudes and perceptions to act as a comparison data set.
Safety is the top priority
Our data showed that the Project Endeavour trials improved perceptions of safety. Our before and after data showed a positive change for participants in their perceptions of safety – before the trial, 68.3% believed self-driving cars are safe, yet this increased to 83.6% afterwards. In our interviews, the words “cautious” and “measured” were often used by participants to describe how the vehicle behaved.
These findings echoed our national survey, which showed that the safety case for self-driving vehicles has yet to fully convince the public. We found that three in ten (29.9%) believe that self-driving vehicles will be safer than traditional vehicles, while 44.2% disagree. A quarter (25.9%) are undecided, but it is plausible that this group is primed to be convinced through trials, for example, or through education and more information. Safety perceptions differed by age. Older people were less positive about self-driving cars and more concerned about safety.
Does the public trust self-driving vehicles?
Cities and roads are complex, and to operate in them effectively, vehicles will need to be able to make decisions rapidly and without human intervention. Trust is a therefore a critical factor that drives acceptance and adoption. Our study showed that perceptions of trust improved for those who participated in the trial: 71.3% believed that self-driving vehicles were trustworthy before the trial experience, this increased to 87.2% afterwards.
We also developed a virtual reality (VR) experience to reach a wider audience with the experience, which showed participants a 3D video of the Oxford trial. In this realistic virtual environment, participants could ‘sit in a car’ and experience the journey themselves. We then ran a post-experience survey to measure perceptions for those who participated virtually.
VR participants were positive, and whilst it was a cross-sectional “moment-in-time” survey, qualitative data highlighted the benefits of VR, including the flexibility to use it at home. We were also able to target a wider community of participants, including elderly people and those with disabilities.
During the pandemic, the team quickly developed a virtual reality driving experience to continue the trial
What’s next for AV technology?
For cities, there is great potential in autonomous vehicle technology in terms of achieving safety and accessibility benefits, and improving the sustainability of transport networks. Our trials highlight how far the technology has come, but also how much further is left to go. The public has mixed views of the future, many still see these vehicles as a novelty, even though autonomy features increasingly in driving assistance technology, such as ABS (anti-lock brake system) that features in many vehicles on the road today.
However, Project Endeavour has demonstrated the importance of trials and user research for understanding the public’s perceptions and views. For another of our projects, D-Risk we went out and spoke to members of the public about their ‘edge case’ driving experiences and gained a unique insight into how people feel about self-driving technology in different parts of the country. Both of these exercises demonstrate the value of engagement – in the design of new services and technologies, the public have a vital role to play. Without building trust and understanding, future technologies could suffer as they may not reflect the needs and preferences of their users – ultimately, the public will influence the wider adoption of any innovation. Deep research exploring AVs, with a focus on safety, is also critical as it will play a central role in shifting people’s views and enabling the development of evidence-based policy – and ultimately, encouraging public acceptance of a technology with huge potential.