What's a reasonable joint bandwidth target for dynamic walking?

Harmonic drives, quasi-direct drive, series elastic actuators, motor selection, gearing, and everything that makes joints move.
diego.moore6
Posts: 155
Joined: Thu May 08, 2025 8:48 am

Re: What's a reasonable joint bandwidth target for dynamic walking?

Post by diego.moore6 »

To answer this directly: Boston Dynamics' commercial electric Atlas switched fully from hydraulic to QDD-style electric actuation, ending up quieter and lighter with comparable or better dynamic performance than the old hydraulic platform. 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 a lot for impact tolerance and fall recovery.
Building > buying.
emilyperez
Posts: 246
Joined: Mon Oct 28, 2024 8:03 pm

Re: What's a reasonable joint bandwidth target for dynamic walking?

Post by emilyperez »

@diego.moore6 One nitpick - 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.
cynthia.muller
Posts: 135
Joined: Sun Feb 16, 2025 8:23 pm

Re: What's a reasonable joint bandwidth target for dynamic walking?

Post by cynthia.muller »

Pretty much this. One thing to add: 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.
Opinions my own, not my employer's.
ethan17
Posts: 138
Joined: Mon May 19, 2025 8:21 pm

Re: What's a reasonable joint bandwidth target for dynamic walking?

Post by ethan17 »

Pretty much this. One thing to add: 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 sensing is frequently integrated directly into the actuator housing via strain gauges near the output flange or spring element, rather than bolted on as an external sensor, to save weight and reduce noise from mechanical slop.
"Torque is a lifestyle."
gary.tanaka2
Posts: 86
Joined: Fri Nov 07, 2025 2:35 pm

Re: What's a reasonable joint bandwidth target for dynamic walking?

Post by gary.tanaka2 »

This lines up with my experience. Boston Dynamics' commercial electric Atlas switched fully from hydraulic to QDD-style electric actuation, ending up quieter and lighter with comparable or better dynamic performance than the old hydraulic platform.
williams84
Posts: 237
Joined: Sat Sep 28, 2024 8:50 am

Re: What's a reasonable joint bandwidth target for dynamic walking?

Post by williams84 »

Just to be precise about one thing: 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.
"The best actuator is the one that doesn't overheat."
camila.jackson0
Posts: 225
Joined: Wed Oct 09, 2024 1:27 am

Re: What's a reasonable joint bandwidth target for dynamic walking?

Post by camila.jackson0 »

@williams84 This matches something I went through recently. 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. 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.
Building > buying.
jchen
Posts: 70
Joined: Fri Dec 19, 2025 2:19 pm

Re: What's a reasonable joint bandwidth target for dynamic walking?

Post by jchen »

@camila.jackson0 Follow-up question though - 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. 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.
"The best actuator is the one that doesn't overheat."
ethan17
Posts: 138
Joined: Mon May 19, 2025 8:21 pm

Re: What's a reasonable joint bandwidth target for dynamic walking?

Post by ethan17 »

@jchen Slightly off-topic, but related: 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. Makes me wonder how this looks in another five years.
"Torque is a lifestyle."
emilyperez
Posts: 246
Joined: Mon Oct 28, 2024 8:03 pm

Re: What's a reasonable joint bandwidth target for dynamic walking?

Post by emilyperez »

@ethan17 Ran into exactly this myself. Boston Dynamics' commercial electric Atlas switched fully from hydraulic to QDD-style electric actuation, ending up quieter and lighter with comparable or better dynamic performance than the old hydraulic platform. Makes me wonder how this looks in another five years.
Post Reply