Graceful Robot Learns to Play a Mean Game of Jenga

Alberto Rodriguez, assistant professor of mechanical engineering at MIT along with a team from the MIT MCube Lab developed a very precisely moving robot that uses machine learning hardware to play a mean game of Jenga. The robot gracefully removes a wooden piece in two motions and the place it at the top of the pile and is able to repeat this motion several times while learning and retaining information about the environment in real time.

The robot efficiently and clearly identifies when a piece feels stuck or free and decides how to extract it using far less data. This approach as a successful example of AI moving into the physical world. The robot learns as it interacts with its environment and captures some of the essential skills that define human manipulation.

Machine Learning Robot Plays Jenga

via MIT News