Humanoid-universe.com

<t>I'll believe the stronger version of that claim when it's independently verified.

Quasi-direct drive (QDD) actuators use lower gear ratios (roughly 6:1 to 10:1) paired with high-torque-density motors, trading some peak force density for backdrivability and low reflected inertia, which matters ...

<t>@carol38 Agreed, and I'd add:

It's common for a single humanoid to mix actuator types by joint - harmonic drives or hybrid geared actuators at the hips and shoulders where sustained torque matters most, and lower-ratio QDD-style actuators at ankles and knees where backdrivability and impact ...

<t>@hill23 Slight correction, though the overall point stands:

Force/torque sensors near the ankle give a direct read on ground reaction forces, which is valuable for balance control, but they add cost, a failure point, and routing complexity right at a joint that already takes the most ...

<t>@barbara50 Just to be precise about one thing:

Proprioception (the robot's sense of its own joint angles, velocities, and forces) tends to get less attention than flashy vision systems, even though a lot of balance and manipulation failures trace back to proprioceptive noise or miscalibration ...

<t>Minor factual note:

Physical Intelligence's pi0 pairs a smaller pretrained vision-language backbone with a separate flow-matching 'action expert' module, which is one way to get fast, high-frequency action output without needing the whole giant language model to run at control-loop speed ...

<t>@barbara_liu +1 to this. Worth adding:

Underactuated hands (fewer actuators than joints, using mechanical coupling to shape the grasp) are a reasonable engineering compromise for robust power grasps on a budget, but they generally can't do fine in-hand manipulation the way a fully actuated ...

To answer this directly: 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.

This matches something I went through recently. Joining an existing robotics community (FRC, university clubs, maker spaces) before going fully solo is a common and effective way to pick up practical fabrication and control skills faster than working entirely alone from tutorials.

<t>@emma_whit This matches something I went through recently.

A realistic hobbyist bill-of-materials for a small biped balloons fast once you add a real IMU, decent motor controllers, a reasonable battery/BMS setup, and enough compute for onboard control - the headline '$5k' budget assumes a fair ...

<t>@forgecam45 Same conclusion I've come to. Also worth noting:

Companies like Figure, 1X, Apptronik, and Agility Robotics are reportedly pulling senior ROS, controls, and learning engineers directly out of warehouse-automation and industrial-robotics roles - a sign of real talent competition ...