Ivan Hernandez Dalas - Mechatronics

Generalist’s latest embodied foundation model, GEN-1, now supports a broad range of robot end effectors. The company said GEN-1 is compatible with five-fingered hands to specialized tools with new modes of actuation and everything in between. By training GEN-1 to work with these new hands, Generalist said it is demonstrating that a single base AI model can learn sensorimotor policies on robots that transfer across radically different ways of interacting with the physical world. Below, the company shared examples of these hands in action and where it thinks this is headed. Generalizing GEN-1 to many hands GEN-1 is pretrained on Generalist’s in-house robotics dataset, which now spans a wide variety of different end effectors across more than half a million hours of real interaction data. Some end effectors involve new form factors with their own actuation schemes and camera positions, a few inspired by real commercial use cases. Others are off-the-shelf tools, printed...

The Robot Report Podcast · Unlocking the Power of Time Series Databases for Industrial and Robotic Systems In Episode 254 of The Robot Report Podcast, we talk to Doug Pagnutti, developer advocate at Tiger Data. He discusses how time series databases like TimescaleDB are improving industrial automation, robotics, and AI applications. He shares insights about integrating these databases with various sensors, managing data at scale, and optimizing performance both on the cloud and on the edge. Doug Pagnutti of Tiger Data. Pagnutti is an industrial developer advocate at Tiger Data, where he helps teams build real-world industrial systems. He spent 12+ years in oil and gas and manufacturing automation, working on PLCs, robots, and SCADA systems before moving into industrial software development at Dell EMC and VTScada. He bridges the OT/IT gap by helping automation engineers tackle time-series data infrastructure for IIoT deployments at scale.      Show timeline ...

A rendering of the planned NEURA Gym RWTH Aachen. Source: NEURA Robotics People are not the only ones who can benefit from going to the gym. Physical AI developers are increasingly turning to specialized facilities to train models enabling robots to perceive, decide, and act in the real world. NEURA Robotics GmbH this week said it has partnered with RWTH Aachen University to address the scarcity of real-world data for training its “cognitive” robotics. “Unlike large language models, which train on trillions of data points, robots have access to a fraction of that volume,” explained NEURA Robotics. “Simulations alone cannot reproduce the friction, variability, and unpredictability of the real-world environment robots will operate in.” The NEURA Gyms will combine physical training and high-fidelity simulation to create high-quality datasets that directly feed the Neuravese open, cloud-based development platform. “Partners can train and validat...

A miniature robot developed at the University of Basel could help prepare teeth for a crown. Photo: University of Basel, Catherine Weyer. By Angelika Jacobs A routine check-up at the dentist ends with bad news: tooth decay has left a large cavity, and the tooth needs a crown. The treatment requires several follow-up appointments. During the first appointment, the dentist removes the decay, fills the cavity and prepares the tooth for the crown. She then takes an impression and fits a temporary crown. The permanent crown is produced based on the impression and can only be placed at a later appointment. In future, this process could become much faster thanks to a small dental robot developed by researchers at the Department of Biomedical Engineering at the University of Basel. The idea came from researchers at the University of Zurich, who were also involved in the development. The prototype is about the size of a wine cork, measuring just 43 by 26 by 28 millimeters. Its motors and co...

ATOMS said it aims to make gainfully employed robots that bring abundance to their owners and society at large. | Source: ATOMS ATOMS, a robotics startup led by former Uber co-founder and former CEO Travis Kalanick, today said it has  raised $1.7 billion in equity investment. Kalanick said the funding will enable ATOMS to complete completion of the “bits-to-atoms” story arc he started at Uber and continued at CloudKitchens. So far, Kalanick hasn’t given many details about the robots the new company plans to build. He did say that ATOMS will act as an OEM that builds “atoms-based” computers for all major industrial sectors. Also as part of this investment, ATOMS merged its various businesses into one equity structure. Lead investors in the round, a16z and Ben Horowitz, are joining the company’s board. Other equity partners include Bain Capital, Uber, Fifth Wall, Chemistry, A*, K5 Global, Abstract, SV Angel, and Alpha Square Group. Debt partners include B...

Ropedia’s HOMIE is a lightweight device that includes four cameras and a swappable battery module. | Source: Ropedia Ropedia, which is building data infrastructure for physical AI, today said it has raised $22 million in pre-Series A funding. Combined with the company’s seed round, it has raised $30 million. The company  plans to use the funding to scale HOMIE, its wearable , head-mounted device. HOMIE captures first-person human movement, object interaction, and spatial context and then feeds that data into Ropedia’s processing and annotation models. “For this round of fundraising, we’re actively expanding our business and technical networks in the sense that we want to build up our teams in hardware, software, data infrastructure, and model development,” Zhaoxi Chen, co-founder and CEO of Ropedia, told The Robot Report . “The second usage is really going to be expanding our business to North America, especially in the United States, because the majority of our exi...