Please create an account to participate in the Slashdot moderation system

 



Forgot your password?
typodupeerror
×
Robotics AI

Robot Combines Vision and Touch To Learn the Game of Jenga (sciencedaily.com) 21

In the basement of MIT's Building 3, a robot is carefully contemplating its next move. It gently pokes at a tower of blocks, looking for the best block to extract without toppling the tower, in a solitary, slow-moving, yet surprisingly agile game of Jenga. From a report: The robot, developed by MIT engineers, is equipped with a soft-pronged gripper, a force-sensing wrist cuff, and an external camera, all of which it uses to see and feel the tower and its individual blocks. As the robot carefully pushes against a block, a computer takes in visual and tactile feedback from its camera and cuff, and compares these measurements to moves that the robot previously made. It also considers the outcomes of those moves -- specifically, whether a block, in a certain configuration and pushed with a certain amount of force, was successfully extracted or not. In real-time, the robot then "learns" whether to keep pushing or move to a new block, in order to keep the tower from falling.

Details of the Jenga-playing robot are published in the journal Science Robotics. Alberto Rodriguez, the Walter Henry Gale Career Development Assistant Professor in the Department of Mechanical Engineering at MIT, says the robot demonstrates something that's been tricky to attain in previous systems: the ability to quickly learn the best way to carry out a task, not just from visual cues, as it is commonly studied today, but also from tactile, physical interactions.

This discussion has been archived. No new comments can be posted.

Robot Combines Vision and Touch To Learn the Game of Jenga

Comments Filter:
  • Another game playing robot. Welcome to the future - MIT style!
    • by Anonymous Coward

      " equipped with a soft-pronged gripper, a force-sensing wrist cuff, and an external camera, all of which it uses to see and feel the tower " -Shoot, a guy could have a pretty good weekend in Vegas with all that..

    • by AHuxley ( 892839 )
      In the Eastern US states they still get to put together a traditional robot.
      In CA its all about the GUI robot software when learning to code.
  • by fahrbot-bot ( 874524 ) on Friday February 01, 2019 @04:18PM (#58057096)
    ... I tried playing Jenga with my friend's Roomba and it did not go well.
  • by bobstreo ( 1320787 ) on Friday February 01, 2019 @04:21PM (#58057114)

    and set it loose in an auto scrapyard.

  • This is all very interesting, but it means nothing until the robot learns to play Aussie Rules Jenga. And for that, it needs to learn how to drink like an alcoholic camel.

    • No, it doesn't mean anything until it can LOOSE at shot Jenga. When robots can get drunk THEN and only then do we have to worry. Also, we have to feel sorry for our Hungover Robot Overlords.
  • I have a friend, Jason, who is a dancer/coreographer, who built a jenga set out of two by fours. He and his troup would play at renaissance faires using only their feet. Let me know when a robot can do that.

    • Most robots consist of a body bolted to the floor and a single appendage. And it still may be able to accomplish all Jenga tasks (with the proper programming supplied by humans or machine learning). Add a second robot for two appendages, still half as many a single human has. Given, proper Jenga rules permit only one hand at a time the second robot "arm" may not be needed/allowed. What is your point? Is renfaire the basis for robot takeover? "Well shite. Thine robot hast bested me best foot model. Her
  • This is a big deal, if it works reasonably well. Interacting with the real world is much more difficult than interacting with a mathematical abstraction.

    This would require very sensitive movement, plus quite a bit of analysis. It may well beat most human players with better forward planning, but that is just minimax.

    We will start to see a *lot* of robots over the next few years.

  • by Anonymous Coward

    No way will it ever beat this biologically based Jenga master. [youtube.com]

    (I know youtube links are a risky click here. This one's apropos, I promise).

Some people claim that the UNIX learning curve is steep, but at least you only have to climb it once.

Working...