CyberRunner AI Robot Beat Humans Labyrinth Game
Similar to this autonomous Menzi Muck excavator, CyberRunner can autonomously beat humans at a classic labyrinth marble game. For those unfamiliar with the game, players basically steer a marble from a start point to the end point, all the while preventing it from falling into any of the holes that are present on the board.



This AI robot leverages model-based reinforcement learning and its ability to make informed decisions about potentially successful behaviors by planning into the future. It essentially learns by collecting experience while playing the game. Observations are captured and then rewarded based on its performance, all through the camera looking down at the labyrinth. The collected ‘memory is then used for model-based reinforcement learning algorithm, enabling it tp recognize which strategies and behaviors are more promising.

Sale

STERLING Games Large Wooden Labyrinth 13.5' x 13' Marble Ball Tilt Maze Wood Game with 60 Waypoints for 6...

STERLING Games Large Wooden Labyrinth 13.5″ x 13″ Marble Ball Tilt Maze Wood Game with 60 Waypoints for 6…

  • PREMIUM CONSTRUCTION: The wooden labyrinth or labrynth is designed in the US and made with selected wooden material. The board and walls are sturdily…
  • 60 HOLE CHALLENGE: This is the advanced level version of the classic wooden puzzle toy. Measures at 13.5″ by 13″ and with 60 waypoints to navigate…
  • EDUCATIONAL DEVELOPMENT: For younger kids this labyrinth toy can be an educational toy to develop fine motor skills. It is also a brain teaser that…

CyberRunner AI Robot Beat Humans Labyrinth Game

Consequently, the way the robot uses the two motors – its ‘hands’ – to play the game is continuously improved. Importantly, the robot does not stop playing to learn; the algorithm runs concurrently with the robot playing the game. As a result, the robot keeps getting better, run after run,” said ETH Zurich researchers.

[Source]