Sign In

Communications of the ACM

ACM TechNews

Using ML to Solve Engineering Challenges

View as: Print Mobile App Share:
An unmanned aerial vehicle (UAV) in flight.

Designing safe and workable autonomous, electric-powered flight is a complex engineering problem.

Credit: WSU Insider

Washington State University (WSU) and Oregon State University researchers developed and used a machine learning (ML) algorithm to extract five optimal designs from roughly 250,000 possible designs for an electric power system for an autonomous aerial drone.

The algorithm could factor multiple design goals like safety to find optimal solutions, drastically lowering the number of design iterations required.

The algorithm assessed less than 0.05% of the designs, with the entire system evaluated, rather than individual components.

WSU's Jana Doppa said, "Until now, nobody has used machine learning to look at the multi-objective setting where there are also safety constraints."

Added WSU's Syrine Belakaria, the ML algorithm can be employed "for other design optimization problems which have a lot of constraints, and you're looking for very specific designs that are able to satisfy all those constraints."

From WSU Insider
View Full Article


Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA


No entries found