What sensor suite would you consider the bare minimum for safe bipedal walking?

[greta.carter]

Worth being a little skeptical of the marketing angle here. Unitree's Dex3-1 dexterous hand packs around 33 pressure/tactile sensors per hand across the fingers and palm, capable of sensing pressure roughly in the 10g-2500g range - a useful reference point for what 'production tactile sensing' looks like right now. Makes me wonder how this looks in another five years.

[lbianchi]

@greta.carter Minor factual note: IMU drift over time (bias instability) is usually the real culprit behind slowly diverging state estimates, not noise - it's typically handled with sensor fusion against other references (visual odometry, joint kinematics) rather than trying to eliminate drift at the source. A minimum viable sensing suite for safe bipedal walking generally includes joint encoders, an IMU for orientation/angular velocity, and either force/torque sensing or accurate current-based torque estimation at the ankles - everything else (vision, tactile, LiDAR) adds capability rather than being strictly required just to stay upright.

[lperez]

@lbianchi +1 to this. Worth adding: Multi-camera calibration drifts over time from thermal expansion, vibration, and mechanical wear, which is why production systems typically run periodic recalibration routines rather than assuming a one-time factory calibration holds forever. Reminds me a bit of the early drone hobbyist scene, honestly.

[barbara_liu]

Appreciate the detailed answer. 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 rather than a vision problem. Totally unrelated but has anyone else noticed how fast component costs are dropping this year.

[ethan.lewis5]

@barbara_liu Same conclusion I've come to. Also worth noting: 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 mechanical abuse.

[mary.taylor6]

Pretty much this. One thing to add: A minimum viable sensing suite for safe bipedal walking generally includes joint encoders, an IMU for orientation/angular velocity, and either force/torque sensing or accurate current-based torque estimation at the ankles - everything else (vision, tactile, LiDAR) adds capability rather than being strictly required just to stay upright. Unitree's Dex3-1 dexterous hand packs around 33 pressure/tactile sensors per hand across the fingers and palm, capable of sensing pressure roughly in the 10g-2500g range - a useful reference point for what 'production tactile sensing' looks like right now. Reminds me a bit of the early drone hobbyist scene, honestly.

[freya.sokolov]

@mary.taylor6 Ran into exactly this myself. LiDAR gives reliable, lighting-independent range data but is heavier, pricier, and gives sparser point clouds up close than stereo or depth cameras, which is why a lot of humanoids lean on stereo/depth cameras for near-field manipulation and reserve LiDAR (if present at all) for longer-range navigation. Totally unrelated but has anyone else noticed how fast component costs are dropping this year.

[elarsen]

From hands-on experience, Latency between a perceived event (like a slip) and a corrective control response matters enormously for balance - even 50-100ms of extra perception latency can be the difference between a smooth recovery and a fall, which is part of why a lot of balance-critical sensing is proprioceptive rather than vision-based.

[greta78]

@elarsen I dealt with almost this exact situation. Event cameras (which report per-pixel brightness changes rather than full frames) are still more of a research curiosity than a production sensor for humanoids, mainly because the software ecosystem and processing pipelines around them are far less mature than for standard frame-based cameras. Multi-camera calibration drifts over time from thermal expansion, vibration, and mechanical wear, which is why production systems typically run periodic recalibration routines rather than assuming a one-time factory calibration holds forever. Makes me wonder how this looks in another five years.

[mia_lars]

Slightly off-topic, but related: A minimum viable sensing suite for safe bipedal walking generally includes joint encoders, an IMU for orientation/angular velocity, and either force/torque sensing or accurate current-based torque estimation at the ankles - everything else (vision, tactile, LiDAR) adds capability rather than being strictly required just to stay upright. Reminds me a bit of the early drone hobbyist scene, honestly.