Humanoid-universe.com

Ran into this exact question at work this week and wanted a sanity check. 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. 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. Open to being corrected on the specifics.

Same conclusion I've come to. Also worth noting: 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. Depth sensing range and reliability both degrade outdoors in direct sunlight for most structured-light and active stereo cameras, since the ambient IR washes out the projected pattern - it's a real limitation for humanoids intended for anything beyond indoor, controlled environments.

@jhansen Side note that might be relevant: 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.

Pretty much this. One thing to add: Sensor fusion mostly earns its keep by covering for each individual sensor's weaknesses - vision struggles with occlusion and lighting, IMUs drift, force/torque sensors are noisy at low loads - fusing them gives a more robust estimate than any one source alone, independent of raw compute.

@emilyperez Tangent, but worth mentioning: 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.

@rossi30 That's the official framing, at least - reality tends to lag a bit. 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. Vibration is one of the most underrated sources of noisy IMU and tactile readings - mounting matters as much as sensor quality, and a poorly isolated mount can add more noise than the sensor's own datasheet specs would suggest.

@choi98 Ran into exactly this myself. SLAM in a working warehouse is harder than in a controlled lab mainly because the map keeps changing - pallets move, people walk through, lighting shifts near dock doors - so a lot of production systems lean on semi-static maps refreshed periodically rather than pure continuous SLAM. 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.

I can speak to this a bit. Tactile skin arrays have improved a lot, but 'good enough to matter' really depends on the task - coarse contact detection across a large area is fairly mature, while fine, high-resolution force distribution sensing (like a human fingertip) is still the harder problem.

This raises a question for me - 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. Estimating joint torque from motor current draw is cheap and requires no extra sensor, but it's less accurate than a dedicated torque sensor because it doesn't capture friction losses through the gearbox - good enough for coarse control, not always for precise force-controlled tasks.

@williams84 That's the official framing, at least - reality tends to lag a bit. 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.