Which joint's torque limit sets how fast a stock humanoid can carry a load?
research/buildbot-torquewall
Which joint's torque limit caps a stock humanoid's carrying? For the G1 it is the waist pitch: doubling it alone takes the heaviest feasible payload from 4 to 6 kg; the H1 has no single bottleneck.
What it found
The Menagerie G1 and H1 humanoids are re-scored on buildbot-carryphase's prescribed single-support carry step (990 cases: payload, step length, step time) with one joint group's torque limits multiplied by 2 or 4. For the G1 the waist pitch (50 N·m in the model) is the wall: doubling it alone lifts the heaviest feasible payload from 4 to 6 kg and the 4 kg speed from 0.10 to 0.20 m/s, while the arms, legs or ankles alone change nothing; all joints at 4× reach 10 kg. The H1 has no single bottleneck (hip pitch, then elbow): only scaling every joint helps. Best speed stays near 0.2 m/s because the gait is prescribed. Limits are the community Menagerie model's, scale factors are swept what-ifs (no actuator is claimed), payload steps are coarse, and no falls or energy are scored. Not a safety case.
A figure from its own output
Paper, code and results
More in Robots on the building site
- Do open permit records predict how long a permit and the build behind it take, better than each city's own averages?Research programme: LightGBM quantile models beat the per-city, per-type empirical-quantile baseline on pinball loss in 13 of 15 city/target/quantile cells (2024 test year, Chicago, San Francisco, Austin); Chicago's tails are the exception
- How long a step can a stock Unitree G1 or H1 take carrying lumber, and does that beat a labourer's carry rate?Can a stock humanoid carry a stud faster than a labourer? A MuJoCo single-support step envelope: the G1 steps 0.50 m with one 2x4 and cannot carry two; the H1 steps 0.20 m with one and 0.10 m with 15 kg; neither beats a labourer who takes a bundle of four, and the model is restrictive (no falls, drops or energy scored).
- Does planning the centre of pressure off the middle of the sole let a stock humanoid carry faster?Does moving the planned centre of pressure along the sole raise a humanoid's carry speed? Not by a constant offset: the G1's 4 kg result moves one grid cell (0.10 to 0.13 m/s) and joint torques, not the sole, set the limit.
- Can a stock Unitree G1 or H1 humanoid carry a stud across a site floor as fast as a labourer?Can a stock Unitree G1 or H1 carry a stud as fast as a labourer? A pendulum gait with no trunk compensation gets the G1 to 0.25 m/s unloaded and 0.10 m/s at 4 kg; a robot with one piece per trip would need 4-13 m/s to match a labourer carrying three.
- Does sweeping the planned centre of pressure from heel to toe let a stock humanoid carry faster?Does a heel-to-toe sweep of the planned centre of pressure raise a humanoid's carry speed? No: one grid cell for the G1 at 4 kg and the unloaded H1, and nothing for heavier loads, because joint torques set the limit.
- Can a stock Unitree G1 or H1 humanoid hold one end of a sheet while a person lifts the other?Can a stock Unitree G1 or H1 hold one end of a sheet while a person lifts the other? For a 22 kg sheet the person carries 11.1-11.5 kg-equivalent in a 2-4 s lift; the G1 manages up to 8 kg, the H1 a 23 kg drywall sheet; lag sets the tilt (2-6 degrees at 0.2-0.5 s).
All 27 Robots on the building site projects →
Other research topics
Proper scoring & elicitation28Effort & peer prediction9Verification games & slashing19Detection & tolerances14Markets as routers13Prediction-market design9Paying contributors8Decentralised training systems23Digital twins & sim-to-real164LLM chip design loops40Robot manufacturing loops53Ambient AI assistant46Self-replicating manufacturing34Simulation-first design64Engineered clothing45Task-specific AI accelerators40Algorithmic chip-design loops42Open-source purpose-built robots44The software-engineering loop43Adversarial human imitation41Open-source residential construction41Autonomous multi-angle capture34Sensing, imaging and 3D31Energy supply and dense energy storage39Robot societies among people31Agentic co-design of low-cost robots27Laundry-hanging robots25Machine learning on qualitative data23Robots on the building site27Civilizations & group psychology1History & human sciences1
Produced with AI assistance (Claude). Results are from the code in this repository and are simulated or stylised unless the abstract says otherwise; nothing here implies affiliation with or endorsement by any lab, researcher or company cited. MIT licensed.