Sign In

Communications of the ACM

ACM TechNews

Making Self-Driving Cars Human-Friendly


View as: Print Mobile App Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook
An autonomous vehicle "senses" pedestrians ahead.

How much will the performance of self-driving cars need to be reduced, or otherwise modified, to enable them to share the roads with pedestrians?

Credit: Bryan Christie Design

Researchers at the U.K.'s University of Leeds have developed a drift diffusion model that could make self-driving vehicles safer for pedestrians by helping to predict when people will cross the road.

The model was tested in different scenarios using the university's HIKER (Highly Immersive Kinematic Experimental Research) pedestrian simulator.

The researchers found that participants used sensory data from vehicle distance, speed, and acceleration and communicative cues to determine when to cross.

Leeds' Gustav Markkula said, "Pedestrians will often feel quite uncertain about whether the car is actually yielding, and will often end up waiting until the car has almost come to a full stop before starting to cross. Our model clearly shows this state of uncertainty borne out, meaning it can be used to help design how automated vehicles behave around pedestrians in order to limit uncertainty, which in turn can improve both traffic safety and traffic flow."

From University of Leeds (U.K.)
View Full Article

 

Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA


 

No entries found