Communications of the ACM
Quicker Eye for Robotics to Help in Our Cluttered, Human Environments
University of Michigan researchers have developed an algorithm that lets machines perceive their environments orders of magnitude faster than similar previous approaches.
Credit: Joseph Xu/Michigan Engineering
University of Michigan (UM) researchers have developed an algorithm that permits machines to sense their surroundings faster, which could help assistive robots operate more efficiently in unstructured environments.
The Pull Message Passing for Nonparametric Belief Propagation algorithm can precisely calculate an object's pose, or position and orientation, within 10 minutes, while previous approaches require more than 90 minutes.
The algorithm is designed to use "pull messaging" to condense the clutter of data-intensive messages about an object's various elements and relationships with each other into a concise dialogue between those elements.
UM's Karthik Desingh said, "We want to scale our work up to multiple objects and tracking them during action execution, and even if the robot is not currently looking at an object. Then, the robot can use this ability to continually observe the world for goal-oriented manipulation and successfully complete tasks."
From University of Michigan News
View Full Article
Abstracts Copyright © 2019 SmithBucklin, Washington, DC, USA
No entries found