Skip to main content

MIT researchers are training a robot arm to play Jenga

Turns out training a robotic arm to play Jenga is a surprisingly complex task. There are, so to speak, a lot of moving parts. Researchers at MIT are putting a modified ABB IRB 120 to work with the familiar tabletop game, utilizing a soft gripper, force-sensing wrist joint and external camera to design a bot that can remove a block without toppling the tower.

The robot was trained with 300 attempts, rather than the thousands it would traditionally take, learning to cluster different attempts into groups as a kind of shorthand similar to how humans teach themselves. With each attempt the robot is pushing against the block, testing for tactile feedback to determine whether it’s a safe bet.

“Unlike in more purely cognitive tasks or games such as chess or Go, playing the game of Jenga also requires mastery of physical skills such as probing, pushing, pulling, placing and aligning pieces. It requires interactive perception and manipulation, where you have to go and touch the tower to learn how and when to move blocks,” says MIT assistant professor Alberto Rodriguez. “This is very difficult to simulate, so the robot has to learn in the real world, by interacting with the real Jenga tower. The key challenge is to learn from a relatively small number of experiments by exploiting common sense about objects and physics.”

The robot has gotten pretty good at making attempts, but the team is quick to note that it’s not quite ready to take on experienced players. Among other things, the robot isn’t able to determine strategic blocks that can sabotage the tower’s strength for upcoming turns. 



from TechCrunch https://tcrn.ch/2Usg4Lt
via IFTTT

Comments

Popular posts from this blog

Apple’s AI Push: Everything We Know About Apple Intelligence So Far

Apple’s WWDC 2025 confirmed what many suspected: Apple is finally making a serious leap into artificial intelligence. Dubbed “Apple Intelligence,” the suite of AI-powered tools, enhancements, and integrations marks the company’s biggest software evolution in a decade. But unlike competitors racing to plug AI into everything, Apple is taking a slower, more deliberate approach — one rooted in privacy, on-device processing, and ecosystem synergy. If you’re wondering what Apple Intelligence actually is, how it works, and what it means for your iPhone, iPad, or Mac, you’re in the right place. This article breaks it all down.   What Is Apple Intelligence? Let’s get the terminology clear first. Apple Intelligence isn’t a product — it’s a platform. It’s not just a chatbot. It’s a system-wide integration of generative AI, machine learning, and personal context awareness, embedded across Apple’s OS platforms. Think of it as a foundational AI layer stitched into iOS 18, iPadOS 18, and m...

The Silent Revolution of On-Device AI: Why the Cloud Is No Longer King

Introduction For years, artificial intelligence has meant one thing: the cloud. Whether you’re asking ChatGPT a question, editing a photo with AI tools, or getting recommendations on Netflix — those decisions happen on distant servers, not your device. But that’s changing. Thanks to major advances in silicon, model compression, and memory architecture, AI is quietly migrating from giant data centres to the palm of your hand. Your phone, your laptop, your smartwatch — all are becoming AI engines in their own right. It’s a shift that redefines not just how AI works, but who controls it, how private it is, and what it can do for you. This article explores the rise of on-device AI — how it works, why it matters, and why the cloud’s days as the centre of the AI universe might be numbered. What Is On-Device AI? On-device AI refers to machine learning models that run locally on your smartphone, tablet, laptop, or edge device — without needing constant access to the cloud. In practi...

Max Q: Psyche(d)

In this issue: SpaceX launches NASA asteroid mission, news from Relativity Space and more. © 2023 TechCrunch. All rights reserved. For personal use only. from TechCrunch https://ift.tt/h6Kjrde via IFTTT