Humanoid-universe.com

<t>Here's the relevant bit as far as I understand it:

Sim-to-real transfer still commonly breaks on contact dynamics - friction, restitution, and deformable/compliant surfaces are the hardest things to model accurately in simulation, so policies trained purely in sim often need real-world fine ...

<t>@greta.carter Thanks for laying this out, genuinely useful.

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>Speaking from personal experience here,

Active employers hiring specifically for humanoid-relevant roles span both established robotics companies (Shadow Robot, Engineered Arts, PAL Robotics) and the newer venture-funded players (Figure, 1X, Apptronik, Sanctuary AI, Boston Dynamics' electric ...

<t>@gary.tanaka2 This lines up with my experience.

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 ...

<t>@sven.wilson4 Genuinely curious -

Multi-camera calibration drifts over time from thermal expansion, vibration, and mechanical wear, which is why production systems typically run periodic recalibration routines rather than assuming a one-time factory calibration holds forever. Sensor fusion ...

<t>@barbara50 One nitpick -

Unitree's Dex3-1 dexterous hand packs around 33 pressure/tactile sensors per hand across the fingers and palm, capable of sensing pressure roughly in the 10g-2500g range - a useful reference point for what 'production tactile sensing' looks like right now. A minimum ...

<t>@garcia51 Slight correction, though the overall point stands:

Diffusion policies model the distribution of possible actions and sample from it, which handles multimodal manipulation tasks (multiple valid ways to grasp something) more naturally than a single deterministic action output, at the ...

One nitpick - Whole-body control (WBC) formulates locomotion and manipulation as a single optimization problem across all joints simultaneously, respecting contact constraints and task priorities - it's more general than ZMP-only approaches but is computationally heavier and harder to tune.

<t>@priya85 From what I've seen:

A strong, well-documented personal project (even a modest DIY build or a solid simulation-based RL project) reportedly carries real weight in hiring for this field, partly because the field is young enough that a demonstrated hands-on track record can meaningfully ...

<t>@lukas.singh1 Tangent, but worth mentioning:

In the UK, base salaries for robotics roles reportedly range from around £50,000 for new graduates up to roughly £200,000 for senior whole-body-control or reinforcement-learning specialists - a notably wide band reflecting how specialized the top ...