Issue 6
7 – 13 Sep 2026
Published 15 Sep 2026 · 23 items

Agility shows $1.8M of revenue, and a robot that draws blood

Robots & HardwareSee all ›
#
8 Sep

Iron Walks Off the Line, XPENG Starts Building Humanoids in Guangzhou (3 minute read)

XPENG's Iron humanoid was finished on a new Guangzhou line and walked off the factory floor under its own power on 7 September, with no remote pilot. The company says more than 80% of the line's fundamental steps are automated, adapted from its electric-vehicle plants to the tighter tolerances humanoids need. Chairman He Xiaopeng said there was no prior playbook when they started developing the lines, and called the result a small first step.

— TechEBlog -humanoidshardware
#
7 Sep

This Robot Will Draw Your Blood Now (5 minute read)

Aletta sweeps the inner elbow with near-infrared light to find a vein, maps its depth with ultrasound, uses Doppler to rule out the artery, then inserts the needle and tips each collection tube exactly nine times. Built by the Dutch firm Vitestro, it is the first autonomous blood-draw device authorised in the United States. The FDA cleared it on 19 August; Vitestro's pitch is doing by algorithm what a phlebotomist does by feel, with no human touching the patient.

— IEEE Spectrumhardwarepolicy-safety
Models & ResearchSee all ›
#
8 Sep

Seeing is Not Believing: Breaking the Physical-to-Digital Trust Boundary in Robotics (paper · 15 minute read)

Changing a single environment variable in ROS 2 lets an adversary run a pre-built hook that intercepts and injects both telemetry and control signals before they are published. A robot can then be hijacked into dangerous behaviour while feeding downstream verifiers synthesised telemetry, and the hooks can travel inside third-party Docker containers. On a physical Franka Emika arm running Secure ROS 2, the fabricated telemetry was injected in real time with about 3 ms of jitter, preserving temporal synchronisation.

— arXivpolicy-safety
#
7 Sep

ContextFlow: In-Context Flow Matching for Robot Manipulation (paper · 15 minute read)

ContextFlow learns robot manipulation from demonstrations held in context, using conditional flow matching rather than the next-token prediction that discretises actions and compounds early errors. Perceiver-style compressors distil visual, proprioceptive and action sequences into compact latents, so several demonstrations can be carried at once. On LIBERO it beats ICRT by 35 percentage points on unseen task configurations and matches a fine-tuned π0 without any fine-tuning of its own.

— arXivmanipulation
#
8 Sep

ETH Zürich's humanoid jumps onto monkey bars and crosses them (video · 3 min)

A humanoid from ETH Zürich's Robotic Systems Lab jumps onto a set of monkey bars, traverses them through sparse bar interactions and lands safely. The lab frames the problem as perceiving thin, overhanging geometry while executing agile, accurate whole-body motions. The video leads IEEE Spectrum's Video Friday for the week.

— YouTube · Robotic Systems Lab: Legged Robotics at ETH Zürichhumanoidslocomotion
#
8 Sep

AgiBot says GE-Act 2.0 keeps improving as its training data scales 100× (6 minute read)

GE-Act 2.0 trains every parameter from random initialisation on embodied manipulation data, rather than adding action prediction to a pretrained video generator. AgiBot reports that scaling the corpus from 300 to 30,000 hours lifted zero-shot success on its G1-OP evaluation from 17.1% to 44.1%, and the number of tasks with any success from 39 to 76, with towel-folding and pen-recapping appearing only at the larger scales. It says performance had not saturated at 30,000 hours.

Data, Sim & ToolsSee all ›
#
9 Sep

MuJoCo 3.13 adds a discrete integrator that keeps stiff springs stable (tool · 2 minute read)

MuJoCo 3.13.0 writes the .mjz root file as model.xml, which survives file renaming, and lets site geometries be associated with meshes. A new discrete integrator merges the constraint solve and the implicit velocity update into one operation, bringing joint, tendon and actuator stiffness into the solver's metric. Passive springs and actuator position gains stay stable at timesteps far beyond the explicit stability limit, and Python 3.15 is now supported, GIL and free-threading alike.

#
7 Sep

Genesis 1.4 exports scenes and trajectories for bit-exact replay (tool · 3 minute read)

Genesis v1.4.0 can export a scene and a complete trajectory as a standalone archive that replays bit-exactly on any machine, aimed squarely at making bug reports reproducible. Only the rigid solver is supported so far, and the static description of entities is now separated from the runtime getters. The breaking changes cover external wrenches applied at any point on a link and morph pose offsets in the velocity and acceleration getters, alongside a long list of asset-parsing fixes.

#
8 Sep

Robotics with Conan: consuming ROS as a regular package (4 minute read)

Conan's team shows how to consume ROS as an ordinary Conan package, rather than installing it system-wide with apt or rosdep on every developer machine and CI agent. It addresses the usual split where C++ control and perception code is managed with Conan while ROS sits separately on top of it. Conan's team has been building the Kilted ROS distribution from source as a regular package for Linux, macOS and Windows, so adding ROS to a C++ project becomes one more requires.

— blog.conan.ioopen-source-tools
Industry & FundingSee all ›
#
7 Sep

Agility Robotics reports $1.8M revenue ahead of humanoid SPAC (3 minute read)

Agility Robotics generated $1.8 million in net sales in 2025 against a $140 million operating loss, according to an S-4 filed ahead of its SPAC merger with Churchill Capital Corp. XI. The deal values Agility at $2.5 billion, roughly 1,400 times annual revenue, with Digit deployed at nine customer sites. The merger is expected to raise more than $620 million, including a $200 million PIPE led by Foxconn.

— The Robot Reporthumanoidsindustry-funding
#
7 Sep

NEURA Robotics and SECO Join Forces to Scale Physical AI from Europe (5 minute read)

NEURA Robotics and SECO will jointly design and manufacture the compute modules for NEURA's cognitive robots, including the 4NE1 humanoid, with series production in Europe. The modules carry Qualcomm Dragonwing processors and form part of NEURA's distributed Brain-plus-Nervous-System architecture. The two will also work together on industrial data and automation for semiconductor and electronics manufacturing.

#
12 Sep

Mecka AI nears $500M valuation in Sequoia-led deal amid rush for robot training data (2 minute read)

Mecka AI, which collects and analyses human motion data to train humanoid and other robots, is nearing a new round led by Sequoia at a valuation of about $500 million, TechCrunch reports, citing two people with knowledge of the deal. It comes three months after a $60 million round led by Framework Ventures; the terms are not final, and neither company commented.