Self-driving cars must reduce traffic fatalities by at least 75 percent to stay on the roads

May 30, 2018, Society for Risk Analysis
Credit: CC0 Public Domain

The race is on for companies to present their driverless cars to the public, but recent collisions involving autonomous vehicles developed by Uber Technologies Inc. and Tesla Inc. have led consumers to questions whether these vehicles can alleviate traffic issues and increase safety. A new study published in Risk Analysis examined the question "How safe is safe enough for self-driving vehicles (SDVs)?"

To answer this question, researchers employed an expressed-preference approach—a method that has not previously been employed in this setting—to determine the socially acceptable risk of SDVs. The results showed that the public will not accept this new technology unless it is shown to be safer, approximately four to five times as safe as human-driven vehicles (HDVs). Despite the conveniences SDVs would bring to individuals, such as the ability to watch a movie, read a book, sleep or surf the internet, the public will be much less likely to accept, or even tolerate, SDVs if they have the same risk level as human driving. As suggested by previous studies, an individual increases his or her demand for when that safety is entrusted to an external factor, such as an automated vehicle.

One of the major motivations behind the development of SDVs is to improve road safety. Human error is believed to cause 94 percent of all crashes in the U.S., and 75 percent in the U.K. While SDVs have the potential to significantly reduce these types of crashes, they also introduce several new road risks, including accidents caused by cyber-attacks. Creating perfectly safe SDVs is both technologically and economically infeasible, but policies can require that the risk of having them on the road be as low as technically achievable.

The study was conducted by Peng Liu and Run Yang, Tianjin University, and Zhigang Xu, of Chang'an University. The survey was distributed to a convenience sample of residents in Tianjin, China. Of the 499 respondents, half were randomly assigned to complete a version of the survey for HDVs, while the other half completed an SDV version. Risk frequencies were expressed as one fatality per a certain number of vehicle-kilometers traveled and as one fatality per a certain number of population, respectively. Respondents were asked to accept or reject each traffic risk scenario at one of four levels: never accept, hard to accept, easy to accept and fully accept.

The results show that the respondents believe that SDVs should be four to five times as safe as HDVs. Current global traffic fatal risk is estimated at 17.4 per 100,000, which is 350 times greater than the frequency accepted by 50 percent of the respondents for SDVs. This implies that respondents expect SDVs to improve safety by two orders of magnitude against the current traffic risk.

Based on the results, the researchers propose the following requirements for SDVs based on the tolerability of risk in industrial safety (a concept developed in the health and safety field) in which risks are distinguished by three criteria: unacceptable, tolerable and broadly acceptable. SDVs that are less safe than human drivers would be set as the unacceptable risk criterion. The tolerable risk is that SDVs be four to five times as safe, meaning they should be able to reduce 75-80 percent of current traffic fatalities. The broadly acceptable risk criterion for SDVs is set as two orders of magnitude lower than current global traffic , indicating a hundredfold improvement over current traffic risks, or the same order of magnitude experienced in public transportation modes, such as rail and commercial aviation.

"Our results and method may help government authorities to establish clear safety requirements for regulating SDVs and also help SDV manufacturers find consumers' expectations for SDVs that must be met," states Liu, Ph.D., assistant professor of industrial engineering.

Explore further: Phone-using drivers knowingly ignore the danger

More information: Peng Liu et al, How Safe Is Safe Enough for Self-Driving Vehicles?, Risk Analysis (2018). DOI: 10.1111/risa.13116

Related Stories

Recommended for you

Permanent, wireless self-charging system using NIR band

October 8, 2018

As wearable devices are emerging, there are numerous studies on wireless charging systems. Here, a KAIST research team has developed a permanent, wireless self-charging platform for low-power wearable electronics by converting ...

Facebook launches AI video-calling device 'Portal'

October 8, 2018

Facebook on Monday launched a range of AI-powered video-calling devices, a strategic revolution for the social network giant which is aiming for a slice of the smart speaker market that is currently dominated by Amazon and ...

Artificial enzymes convert solar energy into hydrogen gas

October 4, 2018

In a new scientific article, researchers at Uppsala University describe how, using a completely new method, they have synthesised an artificial enzyme that functions in the metabolism of living cells. These enzymes can utilize ...

1 comment

Adjust slider to filter visible comments by rank

Display comments: newest first

Eikka
not rated yet May 30, 2018
The comparison is a pile of apples against a pile of oranges.

Most drives in the advanced societies can expect their personal risk of injury or death while -they- are driving to be at least 4-5x less than the global average.

Even locally, let's say the US alone, the number of fatalities collect on a small minority of drivers, places, and situations, which means the average driver is much safer in reality than the average deaths per capita would suggest. For instance, if you don't have a habit of driving drunk, your risk of dying in traffic is reduced considerably.

Trying to lower the bar for self-driving vehicles in this way is just plain evil.

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.