Humanoid-universe.com

<t>From what I've seen:

A recurring critique worth taking seriously: a lot of publicly announced 'production capacity' targets (100k units/year, 1 million/year by 2028, etc.) are aspirational manufacturing targets, not confirmed sales or deployment numbers, and the gap between the two has ...

<t>Pretty much this. One thing to add:

Robotics engineers in the US were reportedly seeing $150k-$205k total comp at mid-level and $205k-$300k at senior level in 2026, with humanoid- and foundation-model-specialist roles clearing $280k-$475k - a meaningful premium over general robotics/automation ...

<t>This raises a question for me -

A minimum viable sensing suite for safe bipedal walking generally includes joint encoders, an IMU for orientation/angular velocity, and either force/torque sensing or accurate current-based torque estimation at the ankles - everything else (vision, tactile ...

<t>@thomasmitchell I'd frame this differently.

Compliant wrists that absorb impact during a bad approach or misjudged contact reduce mechanical stress on the whole arm, which matters a lot for long-term reliability even though it's a less visible feature than the hand itself.

Kind of makes me ...

<t>This lines up with my experience.

Model predictive control (MPC) is still very much alive in production humanoids, often working alongside or underneath learned policies - MPC handles short-horizon dynamically-consistent trajectory optimization while learned components handle perception, task ...

<t>@sarahbernard I can speak to this a bit.

Estimated total operational commercial humanoids worldwide sit around 7,000-8,000 units as of mid-2026, broken down roughly as Optimus Gen 3 (1,000+), Unitree (5,500+ cumulative), Agility Digit (~250), Apptronik Apollo (~80), and Figure (~150) - useful ...

Yeah, this tracks with what I've read as well. Payload-to-hand-weight ratio varies a lot across current dexterous hands, and it's a meaningful tradeoff - more DoF and finer sensing generally means more actuators and mass in the hand itself, which eats into the arm's usable payload budget.

<t>Counterpoint:

Estimated total operational commercial humanoids worldwide sit around 7,000-8,000 units as of mid-2026, broken down roughly as Optimus Gen 3 (1,000+), Unitree (5,500+ cumulative), Agility Digit (~250), Apptronik Apollo (~80), and Figure (~150) - useful context for how early-stage ...

<t>Just to be precise about one thing:

Imitation learning from human demonstration video (without robot teleoperation data) is an appealing way to scale up training data cheaply, but it runs into the embodiment gap - human hand kinematics and force profiles don't map directly onto a robot hand's ...

<t>@wei_ross I don't think that's quite right, for what it's worth.

Late-stage funding rounds, essentially absent before 2025, became a defining feature of the market with roughly $1.1 billion raised in 2025 and about $2 billion in 2026 year-to-date - a sign investors increasingly see specific ...