Robotics & Automation News

Where Innovation Meets Imagination

reinforcement learning

Carnegie Mellon engineer reimagines Interstellar’s TARS robot for Moon and Mars exploration

by

A young roboticist’s late-night experiment has evolved into one of the year’s most original “humanoid” (if it can be called that) concepts – a robot that looks more like a collapsing sculpture than a machine, yet somehow walks, rolls, and balances across multiple terrains.

Aditya Sripada, a graduate of Carnegie Mellon University’s Robotics Institute and a senior robotics engineer at Nimble.ai, has developed TARS3D, a small modular robot inspired by NASA’s fictional assistant from Interstellar.

The project, which he describes as “a desire to reconnect with the simple joy of building robots”, has now earned global recognition, being named a finalist for the Mike Stilman Award at the 24th IEEE RAS Humanoids Conference in Seoul – one of the most prestigious events in humanoid research. [Read more…] about Carnegie Mellon engineer reimagines Interstellar’s TARS robot for Moon and Mars exploration

Pioneering roboticist Rodney Brooks says humanoid robots matching human skills is ‘pure fantasy’

by

Rodney Brooks, the renowned roboticist and co-founder of iRobot (maker of the Roomba), has published a lengthy critique of the current humanoid robotics wave, arguing that today’s approaches will not deliver human-level dexterity on useful timelines.

The essay, titled “Why Today’s Humanoids Won’t Learn Dexterity” was posted on his personal blog on September 26, 2025. In it, Brooks says the idea that humanoid robots will be able to do the manual things that humans do any time within decades is “pure fantasy thinking”.

Brooks frames dexterous manipulation as the core requirement for general-purpose humanoids to make economic sense. [Read more…] about Pioneering roboticist Rodney Brooks says humanoid robots matching human skills is ‘pure fantasy’

After doing the laundry, Figure AI humanoid robot loads the dishwasher

by

Helix, humanoid robotics specialist Figure AI’s Vision Language Action (VLA) model, has shown it can “adapt to dramatically different real-world challenges with nothing more than new data”.

After folding laundry and rearranging packages, Helix is now taking on another everyday task: loading a dishwasher.

At first glance, loading a dishwasher sounds like a simple task – just pick up each object and place it in the dishwasher. [Read more…] about After doing the laundry, Figure AI humanoid robot loads the dishwasher

Intrinsic and Google DeepMind unveil AI breakthrough for multi-robot orchestration

by

Intrinsic, a robotics software company owned by Alphabet, parent company of Google, has announced a major advance in artificial intelligence that enables multiple industrial robots to work together in shared environments without colliding.

The development stems from long-term research led by Google DeepMind Robotics – now apparently rebranded as Gemini Robotics – in collaboration with Intrinsic and University College London.

Details of the project were published in Science Robotics under the title “RoboBallet: Planning for Multi-Robot Reaching with Graph Neural Networks and Reinforcement Learning”. [Read more…] about Intrinsic and Google DeepMind unveil AI breakthrough for multi-robot orchestration

GreyOrange teams up with Google Cloud on ‘breakthrough warehouse AI’

by

GreyOrange, a provider of warehouse orchestration software and hardware, has agreed a collaboration with Google Cloud to develop GreyMatter DeepNav, an AI-powered solution for dynamically managing and optimizing autonomous robotic operations at scale.

GreyMatter DeepNav will bring unprecedented levels of automation, efficiency, and accuracy to warehouse management for retail, logistics, and supply chain companies worldwide.

Large warehouses use Autonomous Mobile Robots (AMRs) to streamline and accelerate operations across the work floor, from receiving to storage, picking, packing, and shipping. But despite the clear benefits of AMRs, many companies struggle with the lengthy ramp-up time of new deployments. [Read more…] about GreyOrange teams up with Google Cloud on ‘breakthrough warehouse AI’

Pushing the boundaries of music: A humanoid robot learns to play the drums

by

An innovative project is transforming the relationship between AI, robotics, and creativity. At its core is a system that teaches a humanoid robot to play the drums – not through direct programming, but through a simulated learning process inspired by how musicians practice and improve. (Video on LinkedIn.)

Picture a live concert where, alongside human musicians, a humanoid robot takes its place behind the drum kit. Science fiction? Maybe for now — but it could soon become reality.

At the forefront of this breakthrough is Asad Ali Shahid, researcher at SUPSI (IDSIA USI-SUPSI) and lead researcher on the project Robot Drummer: Learning Rhythmic Skills for Humanoid Drumming, developed in collaboration with Politecnico di Milano. [Read more…] about Pushing the boundaries of music: A humanoid robot learns to play the drums

Reinforcement learning in robotics: Robots that learn from experience

by

Reinforcement learning (RL) is transforming the way robots interact with the world.

Unlike traditional programming or supervised learning, which depend on pre-defined rules or labeled datasets, RL enables robots to learn through trial and error – much like how humans and animals acquire new skills.

This approach is increasingly vital as robots are deployed in complex, unstructured environments where adaptability is key. [Read more…] about Reinforcement learning in robotics: Robots that learn from experience

Google DeepMind releases new world model with ‘vast space’ to train robots and autonomous systems

by

Google DeepMind has unveiled Genie 3, the latest version of its generative world model, describing it as a step forward in creating more intelligent and adaptable simulation environments – including those for robotics applications.

Although the company’s announcement focuses heavily on advancements in video prediction and general AI learning, Genie 3 has direct relevance for robotics developers who rely on high-fidelity, physics-aware simulation systems to train and test autonomous systems.

Described as a “world model,” Genie 3 can generate realistic interactive environments directly from a single image prompt and a short textual description, such as “driving down a highway at sunset” or “walking in a rainy city”. [Read more…] about Google DeepMind releases new world model with ‘vast space’ to train robots and autonomous systems

The Construct launches ‘intensive hands-on training’ for Unitree humanoid robot

by

The Construct, a Spanish robotics education company celebrated for its innovative, browser-based simulation platform, is bringing theory to life with a new, intensive training course focused on programming the Unitree Robotics G1 humanoid robot.

The exclusive three-day, onsite event will be held in Barcelona from October 15-17, 2025.

Known for making complex robotics development more accessible through its online simulation environment, The Construct is now offering a rare opportunity for direct, hands-on experience with one of the industry’s emerging humanoid platforms. [Read more…] about The Construct launches ‘intensive hands-on training’ for Unitree humanoid robot

Webcast: CogitAI launches ‘industry’s first self-learning, actionable artificial Intelligence’

by

CogitAI has launched what it says is the “industry’s first self-learning, actionable artificial Intelligence”. (See interview with CogitAI CEO below.)

The company says it is “delivering on the future promise of AI”. CogitAI says its Continua software as a service self-learning AI platform.

The Continua platform has the power to turn any process, system, software bot, or real robot into a self-learning autonomous service to drive actionable business outcomes, says CogitAI.  [Read more…] about Webcast: CogitAI launches ‘industry’s first self-learning, actionable artificial Intelligence’