Communications of the ACM
UC Engineers Develop Navigation to Avoid Collisions
The researchers tested the new system in simulations of a virtual supermarket, where multiple autonomous robots must safely navigate aisles to help shoppers and employees.
Credit: Shutterstock
A collision-avoidance system developed by the University of Cincinnati (UC)'s Daegyun Choi and Anirudh Chhabra brings engineers a step closer to designing robots to repair satellites or spacecraft in orbit.
Choi said the system uses explainable artificial intelligence, which employs fuzzy logic that relies on gradations of truth rather than a binary right or wrong, and which enables engineers to understand input-output relationships through observed rules.
The system prototype was tested in a two-dimensional virtual supermarket where autonomous robots had to navigate aisles to help shoppers and workers.
The collision-avoidance system also may be used for drones or flying cars, as it also performs well in three-dimensional space.
From UC News
View Full Article
Abstracts Copyright © 2022 SmithBucklin, Washington, DC, USA
No entries found