Sign In

Communications of the ACM

ACM TechNews

Amarsi Project Could See Robots Learn From Co-Workers

View as: Print Mobile App Share:


The Adaptive Modular Architecture for Rich Motor Skills (AMARSi) project aims to build humanoid robots that can autonomously learn and develop motor skills in open-ended environments by learning from the data provided by movement and rewiring their circuits to process and store the new knowledge they have acquired. Technology supporting the robots includes dynamic neural networks, new robotics hardware designs, and complex software algorithms.

AMARSi relies on a biologically inspired view of motor skills that goes beyond traditional robotic designs, says project coordinator Jochen Steil. AMARSi researchers hope their architecture will enable robots to learn by interaction, which involves a combination of kinesthetic learning, imitation, and exploration.

To develop advanced, autonomous robotic systems, scientists need to both reverse and forward engineer biological systems, says University of Washington research scientist Payman Arabshahi.

View Full Article


Abstracts Copyright © 2010 Information Inc., Bethesda, Maryland, USA


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account