How do commercial humanoids handle actuator failure gracefully (or don't they)?

Harmonic drives, quasi-direct drive, series elastic actuators, motor selection, gearing, and everything that makes joints move.
scott21
Posts: 256
Joined: Fri Oct 11, 2024 9:57 am

Re: How do commercial humanoids handle actuator failure gracefully (or don't they)?

Post by scott21 »

Tangent, but worth mentioning: Cycloidal drives are a less common but real alternative to harmonic drives - lower cost at high ratios, decent efficiency, but historically bulkier for the same torque density, which is part of why they haven't fully displaced harmonic drives in commercial hips and knees. Anti-backlash techniques in production actuators range from mechanical preloading (spring-loaded gear meshes) to purely software compensation that models and corrects for known backlash in the control loop.
she/her
emilyperez
Posts: 246
Joined: Mon Oct 28, 2024 8:03 pm

Re: How do commercial humanoids handle actuator failure gracefully (or don't they)?

Post by emilyperez »

This is a great summary, thanks. 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. 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.
young58
Posts: 106
Joined: Fri Oct 17, 2025 6:16 am

Re: How do commercial humanoids handle actuator failure gracefully (or don't they)?

Post by young58 »

I dealt with almost this exact situation. 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. Cycloidal drives are a less common but real alternative to harmonic drives - lower cost at high ratios, decent efficiency, but historically bulkier for the same torque density, which is part of why they haven't fully displaced harmonic drives in commercial hips and knees.
they/them
charlesbianchi
Posts: 157
Joined: Fri Apr 18, 2025 2:51 am

Re: How do commercial humanoids handle actuator failure gracefully (or don't they)?

Post by charlesbianchi »

Just to be precise about one thing: 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.
she/her
freya.sokolov
Posts: 60
Joined: Sun Mar 08, 2026 12:23 pm

Re: How do commercial humanoids handle actuator failure gracefully (or don't they)?

Post by freya.sokolov »

Related question - Cycloidal drives are a less common but real alternative to harmonic drives - lower cost at high ratios, decent efficiency, but historically bulkier for the same torque density, which is part of why they haven't fully displaced harmonic drives in commercial hips and knees. Reminds me a bit of the early drone hobbyist scene, honestly.
Ex-automotive, now full-time robots.
barbara.jones
Posts: 164
Joined: Fri Feb 28, 2025 6:12 pm

Re: How do commercial humanoids handle actuator failure gracefully (or don't they)?

Post by barbara.jones »

@freya.sokolov One nitpick - 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.
he/him | robotics hobbyist since the DARPA Grand Challenge days
byang
Posts: 142
Joined: Thu Apr 17, 2025 5:26 am

Re: How do commercial humanoids handle actuator failure gracefully (or don't they)?

Post by byang »

@barbara.jones I'll believe the stronger version of that claim when it's independently verified. Anti-backlash techniques in production actuators range from mechanical preloading (spring-loaded gear meshes) to purely software compensation that models and corrects for known backlash in the control loop.
"Torque is a lifestyle."
servobre20
Posts: 58
Joined: Fri Mar 13, 2026 6:16 am

Re: How do commercial humanoids handle actuator failure gracefully (or don't they)?

Post by servobre20 »

@byang I'd take that specific number with a grain of salt, honestly. For dynamic walking and recovery, joint control bandwidth in the tens of Hz range is typically necessary to react to a stumble before the center of mass gets too far outside the support polygon.
Ex-automotive, now full-time robots.
chloe_jack
Posts: 176
Joined: Sat Nov 30, 2024 12:42 pm

Re: How do commercial humanoids handle actuator failure gracefully (or don't they)?

Post by chloe_jack »

@servobre20 Follow-up question though - 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.
"The best actuator is the one that doesn't overheat."
jonathan.rao1
Posts: 156
Joined: Fri Dec 20, 2024 7:58 pm

Re: How do commercial humanoids handle actuator failure gracefully (or don't they)?

Post by jonathan.rao1 »

@chloe_jack Worth being a little skeptical of the marketing angle here. Series elastic actuators add a spring element in series with the drivetrain, which gives cheap, precise torque sensing (measure spring deflection) and passive shock absorption, at the cost of reduced control bandwidth and added complexity. 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.
Currently: 3D printing my way to bankruptcy.
Post Reply