Self-driving cars are already hitting roads across the country.

But for all their high-tech whositswhatsits, they still lack one important element: social awareness.

While autonomous technologies have improved substantially, they still ultimately view the drivers around them as obstacles made up of ones and zeros, rather than human beings with specific intentions, motivations, and personalities, according to MIT CSAIL.

The Computer Science and Artificial Intelligence Laboratory has been exploring whether self-driving vehicles can be programmed to classify motorists social personalities and better predict what other cars will do.

Using a psychology tool called Social Value Orientation (SVO), scientists classified driving behavior based on the degree to which someone is selfish (egoistic) versus altruistic or cooperative (prosocial).

The system estimates motorists SVOs to create real-time trajectories for autonomous cars.

Working with and around humans means figuring out their intentions to better understand their behavior, graduate student Wilko Schwarting, lead author on the paper, said in a statement.

Peoples tendencies to be collaborative or competitive often spills over into how they behave as drivers, he continued. In this paper, we sought to understand if this was something we could actually quantify.

Testing their algorithm on cerebral tasks like merging lanes and making unprotected left turns, the team showed they could better predict the behavior of other cars by a factor of 25 percent.

In left-turn simulations, for example, their car knew to wait when the approaching vehicle had a more egoistic driver and to make the turn when the opposing motorist was more prosocial.

Existing technology can warn operators of oncoming traffic or blind-spot automobiles. But CSAILs platform takes it a step further by, for instance, providing a warning in the rear-view mirror that an oncoming car has an aggressive driver.

Creating more human-like behavior in autonomous vehicles is fundamental for the safety of passengers and surrounding vehicles, since behaving in a predictable manner enables humans to understand and appropriately respond to the AVs actions, Schwarting explained.

Moving forward, the team plans to apply their model to pedestrians, bicycles, and other components of driving environments, as well as robots that regularly interact with humans.

The MIT CSAIL systemnot yet robust enough to be implemented on real roadsis described in full in a paper published this week by the journal Proceedings of the National Academy of Sciences.

More on Geek.com:

Continue reading here:
MIT Researchers Teach Autonomous Cars to Predict Driver Behavior - Geek