Multi-floor SLAM and a blocked robot deal
StrataSLAM keeps one map per floor in ROS 2, China blocks SoftBank's ABB Robotics purchase, and Meituan releases a 15,060 hour desktop VLA model.
#robot-learning
13 posts
Robot ID cards, weight space and drone noise
China drafts a lifetime ID code for every robot it builds, ros2_control ships Lyrical Luth breaking changes, and drone noise reaches the zoning board.
Spot for ICE, Digit at GXO, ROS 2 agents
Boston Dynamics staff fail to stop an ICE Spot sale, Agility adds two GXO sites, Bedrock raises $600M, and a free ROS 2 to VLA course ships its code.
End Effector Poses Buy Generality, Force Buys Throughput
Dream-1 and Eka's VFA model both claim general manipulation, but their action spaces differ, and that choice decides whether a robot can act on contact.
Dream-1, Alpamayo and robot data engines
Dream Labs opens a zero-shot robot API, Bolt picks NVIDIA Alpamayo for 25,000 Lucids, NVIDIA opens the GR00T data engine, and RoboStack merges four ROS repos.
Optimus line, VFA funding, lab agents
Tesla starts Optimus production in Fremont, Eka raises $150M for a vision-force-action model, AGCO signs Digit, and a benchmark scores agents on wet lab work.
Dream API, lab benchmarks, biometric suits
Dream Labs opens a $3/hour robot API trained on video, Agility's co-founder questions Digit 5, and delivery robots face Illinois biometric suits.
robot-dataphysical-airobot-learningdata-labelingsimulationfine-tuningrobot-foundation-modelsbenchmark-evaluation
The Robot Data Refinery Is the Real Product
Google, Physical Intelligence and Humxn all launched on 13 October. What they actually sell is the machinery that turns experience into training data.
World-Action Models Agree on the Bottleneck, Not the Fix
A vendor launch and four arXiv papers attacked the robot data bottleneck in one week, and disagree about where a policy's generality comes from.
The Robot Stack Is Going Modular, and the Guard Layer Has No Home
HomeBody deletes the learned policy, FLUX 3 Action shrinks it to a checkpoint, and RoboHarm shows the frontier model that takes over refuses almost nothing.
Open 7B action model, all-linear hand
Black Forest Labs opens a 7B world action model that tops a public robot benchmark, a Duke lab builds a hand from linear motion, and HFlow opens data tooling.
Atlas Metaplant opens, Halos safety stack
Boston Dynamics starts training Atlas inside Hyundai's Metaplant America, NVIDIA argues safety has to be a whole stack, and robot policies rarely refuse.
The Robot Data Bottleneck Has Two Halves, and Only One Scales
Figure's Helix 2.5 shows human video scales robot policy, while four papers published the same week show what vision cannot see. Robot data is splitting in two.