What torque sensor accuracy is actually achievable at low cost?

[matthew.yamamoto0]

@ethan17 This matches what I've seen too. Motor cogging torque (the 'notchy' feel from magnet-to-slot interactions) shows up as small periodic torque ripple that can translate into visible joint jitter at low speeds - it's usually mitigated with skewed magnets, better current control, or software compensation tables.

[sarah.santos3]

@matthew.yamamoto0 Small correction on one detail: 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 tolerance matter more.

[james15]

@sarah.santos3 Just to be precise about one thing: 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. 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 tolerance matter more. Reminds me a bit of the early drone hobbyist scene, honestly.

[ethan.lewis5]

@james15 One nitpick - A lot of the per-joint cost in a modern actuator isn't the motor - it's the combination of a precision gearbox, integrated encoders, torque sensing, and the driver electronics, all of which have to fit inside a housing the size of a fist.

[young56]

From hands-on experience, A lot of the per-joint cost in a modern actuator isn't the motor - it's the combination of a precision gearbox, integrated encoders, torque sensing, and the driver electronics, all of which have to fit inside a housing the size of a fist.

[rao91]

Not to derail, but this reminds me of something adjacent: 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.

[betty.king]

Sorry if this is a basic question, but 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. Backdrivability is the property that lets an external force move a joint without destroying the gearbox or the motor - it's central to both safe human-robot contact and to letting a leg comply naturally when the robot stumbles.

[sven.smith4]

Agreed, and I'd add: A lot of the per-joint cost in a modern actuator isn't the motor - it's the combination of a precision gearbox, integrated encoders, torque sensing, and the driver electronics, all of which have to fit inside a housing the size of a fist. 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.

[harmonicjen60]

Respectfully, I think this undersells it a bit. 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. Motor cogging torque (the 'notchy' feel from magnet-to-slot interactions) shows up as small periodic torque ripple that can translate into visible joint jitter at low speeds - it's usually mitigated with skewed magnets, better current control, or software compensation tables. Reminds me a bit of the early drone hobbyist scene, honestly.

[scott.andersson5]

Yeah, this tracks with what I've read as well. 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. Backdrivability is the property that lets an external force move a joint without destroying the gearbox or the motor - it's central to both safe human-robot contact and to letting a leg comply naturally when the robot stumbles.