Doing Recursive tasks using human level artificial intelligence

Note:  To make this website free to the public please click on an ad to support my sponsors or you can make a tax-deductable donation using Paypal (click on the donation icon on the left).

Building a robot to play chess is easy. Building a robot to play Zelda is 10 times harder. Even better, is to build a robot that can play any boardgame or any videogame. In fact, building a robot that can do any human task is the holy grail of Artificial Intelligence.

In my patent applications and copyrights (filed starting in 2006), I described a robot that has human level artificial intelligence. The example used in my books and patent applications is playing the legend of zelda. I chose this game because It takes a teenager or adult in order to pass the game. If you ask a 10 year old child to play zelda, they won't be able to pass the game.

The reason RPG games, like Zelda is so difficult is because there are no instructions given to the player. In fact, the rules of the game and the objectives of the game are not disclosed. The robot has to discover the rules and objectives of the game through trial and error.

This robot is playing Zelda for the first time and he has no prior knowledge of this game. He was given vital information through an instruction manual. The instruction manual basically states: your objective is to pass 8 levels in the game to save the princess. The location of level 1 is given to the robot. The instructions stated: the first level is located up north near a bridge. The location of the remaining 7 levels are not disclosed to the robot and he has to search for these levels himself using logic and common sense.

This video is silent because I wanted the viewers to see the thoughts of the robot while playing the game. The robot's conscious is the voice in his head that gives instructions or information. This conscious allows the robot to play the game in an intelligent manner.