Read This

How anthropology can make autonomous cars safer

Engineering and assembling a rolling cocoon is the easy part of creating an autonomous car, relatively speaking. The hard part is dealing with humans – these machines' end users, and pedestrian and vehicular obstacles. We are too complex, diverse, and unpredictable. That's why Nissan has hired Dr. Melissa Cefkin.

A design anthropologist based at the Nissan Technical Center in Silicon Valley, Cefkin is charged with leading the research into the complex web of human-machine interaction: within and outside the vehicle, and in the context of the broader culture. "One of the big questions," Dr. Melissa Cefkin says, "is how many regularities and universalities are there in human behavior, or is there just kind of infinite variation?"

Cefkin's project is to take an empirical look at what happens in shared spaces between and among humans and vehicles to decipher what that says about patterns of interaction. Her goal is, "to help the autonomous vehicle know how to 'behave' properly."

To this end, Cefkin employs all of the core tools of anthropology – individual and group interviews, charted observation, and codified ethnographic videography among them – in order to seek patterns that can be fed into the decision chains of our robot overlords.

Some of Cefkin's more fascinating findings point out the subtlety of the challenges that autonomous vehicles will face.

Because they are designed to maximize efficiency, and lack free will, computerized cars tend to be inherent rule followers. So at highly regulated intersections like stoplights, where nearly everyone abides the rules, they do very well. But at more ambiguous locales, things can get a bit more confounding.

Read the rest at The Drive. This article excerpt is reprinted with permission.

Related Video:

More Information