Communications of the ACM
New Tech Uses Hardware, Software to Train Dogs More Efficiently
A team of researchers at North Carolina State University has assembled a customized suite of technologies that allows a computer to train a dog without human assistance.
Credit: NCSU News
North Carolina State University (NCSU) researchers have developed and used a customized suite of technologies that enables a computer to autonomously train a dog.
"We use sensors in custom dog harnesses to monitor a dog's posture, and the computer reinforces the correct behavior quickly and with near-perfect consistency," says NCSU professor David Roberts.
NCSU professor Alper Bozkurt notes the technology integrates fundamental principles of animal learning into computational systems, and it can probably be applied to a wide range of canine behaviors.
Each harness incorporates a small computer that wirelessly transmits sensor data.
The researchers also created an algorithm that triggered a beeping sound and the release of dog treats from a nearby dispenser whenever the dog's harness sensors detected the canine went from standing to sitting. However, they had to ensure the reinforcement was given shortly after the desired posture was exhibited, and also that rewards were only given for the correct posture. The researchers addressed this by working with 16 volunteers and their dogs to optimize the algorithm, finding the best possible combination of speed and accuracy. They then compared the algorithm's timing and accuracy to that of an expert human trainer, and found the algorithm rewarded the appropriate behavior 96% of the time.
From NCSU News
View Full Article
Abstracts Copyright © 2016 Information Inc., Bethesda, Maryland, USA
No entries found