What happens when humans regain control of driverless vehicles

driving-steering-wheelAs the technological side of driverless technology has become more powerful, we have increasingly turned to more philosophical questions around the technology.  The so called trolley dilemma is one well known example, but we’ve also had questions around how driverless technology will work alongside humans, and studies into how pedestrians may change their behaviors in a driverless world.

A recent Stanford study takes another angle, and examines how we might transition between the car driving and us driving.  How might such a scenario be made as smooth as possible?  The study placed a number of test drivers on a track to find out, and the results suggest the transition isn’t as smooth as we might think.

Taking back control

“Many people have been doing research on paying attention and situation awareness. That’s very important,” the authors say. “But, in addition, there is this physical change and we need to acknowledge that people’s performance might not be at its peak if they haven’t actively been participating in the driving.”

Now, it should be said that the differences in human performance weren’t unsafe or anything, it wasn’t like they were missing red lights or anything of that nature, but as the researchers measured performance, there was a noticeable drop during that transition period as the driver got back into the flow of things again.  The authors believe this could be a time of high risk of accidents.

Participants were asked to drive a simple route on a test track, with the route involving a lane change.  At a particular point, the drivers let the car take over and watched as the car did a few laps itself.  They then retook control from the vehicle and drove some more laps themselves.

The catch was that the autonomous technology may have made a few alterations to the car.  For instance, it might have altered the steering ratio from the normal 15:1 to 2:1, with the humans receiving advanced warning of these changes, and a period of straight driving to familiarize themselves with them.  Despite this however, they did suffer reduced performance when undertaking more challenging maneuvers as a result.

“Even knowing about the change, being able to make a plan and do some explicit motor planning for how to compensate, you still saw a very different steering behavior and compromised performance,” the authors say.

The period of adjustment is due to our implicit motor control taking time to figure out what changes really mean, and it can only really do so through experience.  The experiment was a good example of classic neuroscience experiments to test motor adaption.

The impact on driverless technology

So what impact might this have on our transition to driverless technology?  We should remember that the reductions in our capabilities were only small, and were experienced for a relatively small period of time.  Nonetheless, given the amount of driving people do, the changes are still noteworthy, and even small reductions in effectiveness are important.

The authors believe however that manufacturers need to give some consideration to this handover period, especially as we are likely to have a period whereby humans can easily regain control of their vehicles.  That this experiment showed the reductions in relatively benign conditions merely emphasizes the possible challenges in this area.

“If someone is designing a method for automated vehicle handover, there will need to be detailed research on that specific method,” they conclude. “This study is tip of an iceberg.”

Related

Facebooktwitterredditpinterestlinkedinmail

Leave a Reply

Your email address will not be published. Required fields are marked *

Captcha loading...