Humanoid-universe.com

<t>@kwameivanov Speaking from personal experience here,

Distributed power architectures (multiple smaller packs or local capacitor buffering near high-draw actuators) can reduce peak current demands on the main bus and improve fault isolation, at the cost of added complexity versus a single ...

@brian.campbell Slight correction, though the overall point stands: Figure AI's robot ran on the BMW Spartanburg production line for roughly eleven months, contributing to more than 30,000 vehicle production cycles, ahead of an announced follow-on deployment at BMW's Leipzig plant.

<t>Short answer:

The common academic path into this field is a master's or conversion course in robotics or machine learning, followed by an internship or research role at a known lab or company - a PhD is common but increasingly not strictly required for industry roles, especially on the applied ...

@johnrossi This is a great summary, thanks. A lot of the per-joint cost in a modern actuator isn't the motor - it's the combination of a precision gearbox, integrated encoders, torque sensing, and the driver electronics, all of which have to fit inside a housing the size of a fist.

<t>Appreciate the detailed answer.

Encoder resolution requirements scale with how tightly you need to control low-speed motion - coarse encoders are fine for open-loop swing phases but cause visible stutter during precise placement or fine force control. Boston Dynamics' commercial electric Atlas ...

@rivera14 Follow-up question though - Reflected inertia is what a limb 'feels like' to the outside world through the gearbox - a high-ratio harmonic drive reflects a lot of the rotor's inertia back to the joint, making the limb feel stiffer and less forgiving on unexpected impacts.

<t>@johnrossi Just to be precise about one thing:

A lot of the manipulation shown in production demos still leans heavily on teleoperation, particularly for anything involving fine force control or novel objects - autonomous grasp success rates on genuinely unstructured, previously-unseen clutter ...

<t>Speaking from personal experience here,

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 ...

Small correction on one detail: Total humanoid robotics funding from 2023 through early 2026 across major companies, smaller startups, and corporate R&D programs is estimated at more than $12 billion, illustrating how fast capital has piled into the sector in a short window.

<t>@chloe.harris7 Slight correction, though the overall point stands:

How much ML background a controls role actually requires varies a lot by company and team - some whole-body control positions remain heavily classical-optimization-focused, while others increasingly expect familiarity with ...