Humanoid-universe.com

<t>@erik_novi Counterpoint:

Harmonic drives use high reduction ratios (often around 100:1), which gives excellent force density, stiffness, and thermal endurance for continuous-duty lifting - the tradeoff is low backdrivability and some backlash. Motor cogging torque (the 'notchy' feel from ...

<t>Worth being a little skeptical of the marketing angle here.

Isaac Lab (the successor to Isaac Gym) is widely used for large-scale parallel RL training thanks to GPU-accelerated physics, while MuJoCo is often used as a secondary 'sim-to-sim' validation step because its contact dynamics are ...

<t>I'd take that specific number with a grain of salt, honestly.

Cross-embodiment training (training one policy across data from multiple different robot bodies) has shown some real transfer benefits for high-level behaviors, but low-level control (exact joint torques, timing) still tends to need ...

<t>@brian.campbell Agreed, and I'd add:

A BMS (battery management system) has to guard against transient current spikes from sudden gait changes or lifting motions, not just steady-state draw - peak current headroom and fast-acting protection logic matter as much as total capacity for real-world ...

<t>@freya.sokolov Here's the relevant bit as far as I understand it:

Apptronik raised a $520 million Series A at a $5 billion valuation (bringing total funding to roughly $935 million), with its Apollo humanoid reportedly being tested by Mercedes-Benz - a notable data point for how much capital a ...

<t>Respectfully, I think this undersells it 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 ...

<t>Counterpoint:

Wet, oily, or otherwise low-friction objects are still a genuine edge case for most current hands, since tactile sensing and grasp-force controllers are typically tuned and validated on dry, higher-friction test objects. Cable routing through a wrist joint with multiple degrees ...

<t>@young58 I'd push back on this a bit.

Bimanual manipulation - two arms coordinating on one task - is harder than it looks mostly because of the added degrees of freedom and the timing/force coordination required; a lot of 'two-handed' demos are actually closer to two independent single-hand ...

This raises a question for me - 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 central pack.

<t>@servoken70 That's the official framing, at least - reality tends to lag a bit.

The market consolidation question (which of the 100+ companies chasing this space survive) is complicated by the fact that hardware is genuinely capital-intensive to scale, so smaller players without a clear cost ...