Sign In

Communications of the ACM

ACM TechNews

Smart Transportation Systems Need to Reckon with Rogues


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
A drone faces off with a full-size vehicle in an intersection.

Researchers in Arizona State University's Make Simple Programming Laboratory have developed an algorithm to help autonomous vehicles deal with situations in which any car can start accelerating or braking at any point in time relative to transiting an intersection.

Credit: Erika Gronek/ASU

Arizona State University (ASU) engineers have come up with an algorithm that manages autonomous vehicle traffic through intersections safely, under real-life conditions.

Developed by researchers in ASU's Make Programming Simple Laboratory, the robust and resilient intersection management (R2IM) algorithm deals with scenarios in which any car can start accelerating or braking at any time relative to traversing an intersection.

ASU's Mohammad Khayatian said, "Our approach uses a surveillance system, such as cameras mounted at an intersection, that can detect if a vehicle is not following its expected trajectory beyond a tolerance limit. If that's the case, the intersection manager declares that vehicle as 'rogue,'" then alerts all approaching vehicles so they react accordingly.

From ASU Now
View Full Article

 

Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA


 

No entries found