Home / Research / Robotics / Robots that can adapt like animals

Robots that can adapt like animals

An intelligent trial-and-error algorithm is introduced that allows robots to adapt to damage in less than two minutes in large search spaces without requiring self-diagnosis or pre-specified contingency plans. Before the robot is deployed, it uses a novel technique to create a detailed map of the space of high-performing behaviours. This map represents the robot’s prior knowledge about what behaviours it can perform and their value. When the robot is damaged, it uses this prior knowledge to guide a trial-and-error learning algorithm that conducts intelligent experiments to rapidly discover a behaviour that compensates for the damage.

http://www.nature.com/nature/journal/v521/n7553/full/nature14422.html

About Mohammad

Dr Mohammad Khazab completed his Ph.D. in Computer Systems Engineering (Artificial Intelligence) at the University of South Australia in 2011. He has worked as Senior Software Engineer, Web Developer, and Research Associate on various projects. Currently he works at Schneider Electric on the design and development of new software solutions for smart devices used for home automation and Internet of Things. He's also been working on enterprise software for supply chain network simulation and optimisation, advanced planning and scheduling. In his spare times, In his spare times, he works on creating websites and mobile applications (Web2day Design), researching and writing about cutting-edge technologies in this blog. He has ambitions to solve real-world problems, and to use his knowledge and skills to develop useful applications.

Leave a Reply

%d bloggers like this: