OK-Robot is an open framework for zero-shot, natural language pick-and-drop system in arbitrary homes.

Boston Dynamics has released footage showing how its flagship Atlas humanoid robot would cope in a factory environment as it lifts heavy struts and puts them into a flow cart.

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Karel Čapek explains what everyone has been getting wrong about his creation

by Zipeng Fu*, Tony Z. Zhao* and Chelsea Finn at Stanford

Working in a greenhouse is both strenuous and time-consuming. The picking robot from ETH spin-off Floating Robotics takes on particularly repetitive tasks, thereby alleviating the strain on human pickers. It is currently undergoing testing at Beerstecher AG in Hinwil. ETH Social Media Links: Faceb...

In a series of experiments published in Nature Communications, we found evidence that human judgments are indeed systematically influenced by adversarial perturbations.

Self-driving cars could give people who live outside big cities and are unable to drive more mobility, but the technology must still overcome some hurdles

Japanese researchers have created and open-sourced a flying firefighting hose that levitates and steers itself to fight fires using its own water pressure as a two-part propulsion system, spraying water down onto fires and keeping operators safe.

A club for Meccano enthusiasts in London, holding four meetings each year and a public Meccano Show in Falconwood.

This paper introduces Diffusion Policy, a new way of generating robot behavior by representing a robot's visuomotor policy as a conditional denoising diffusion process. We benchmark Diffusion Policy across 12 different tasks from 4 different robot manipulation benchmarks and find that it consistentl...

MIT Professor Justin Solomon applies modern geometric techniques to solve difficult problems in a variety of areas, such as developing machine learning models that perform more accurately on target datasets and helping autonomous vehicles identify pedestrians.

Artificial general intelligence will not arise in systems that only passively receive data. They need to be able to act back on the world.

This book offers a comprehensive introduction to the central ideas that underpin deep learning. It is intended both for newcomers to machine learning and for those already experienced in the field.

The goal of the new research is to help autonomous vehicles make “good” decisions.

Dobb·E is an open-source framework for teaching robots new household tasks in 20 minutes via imitation learning.

Engineers at The University of Manchester have built and flown the world’s largest quadcopter drone.The drone, made from a cardboard-like material called foamboard, measures 6.4m (21 ft) corner to corner and weighs 24.5kg – 0.5kg less than the weight limit set by the Civil Aviation Authority.The inn...

Cambridge scientists have shown that placing physical constraints on an artificially-intelligent system – in much the same way that the human brain has to

傅利叶是谁？他是19世纪伟大的数学家，而21世纪的上海傅利叶智能科技有限公司则是工匠，科学家和艺术家们的组合。最好的公司，不是创始人们想要成立公司，而是创始人们想要改变整个世界。

In the background of image recognition software that can ID our friends on social media and wildflowers in our yard are neural networks, a type of artificial intelligence inspired by how own our brains process data. While neural networks sprint through data, their architecture makes it difficult to ...

AI programs usually master either information-perfect games like chess or information-imperfect games like poker, but "Student of Games" is a general algorithm that can master both types.

In 2021, robotics startups were flying high. Unlike other categories that had buckled under the strains of a global pandemic, interest in automation was

Building AI-powered robots that can learn how to interact with the physical world will enhance all forms of repetitive work.

Can this actually make a difference, though?

Athletic, insect-scale long jumping robots reach where others can't

Reinforcement learning lets 3D printers exploit, not suppress, coiling instabilities.

A group of small, simple robots can make a collective decision by exchanging infrared light signals in a process inspired by how bees decide where to build their nests