Humanoid-universe.com

@emilyperez Side note that might be relevant: Object occlusion by the robot's own hand during the final approach to a grasp is a common and annoying perception problem - the closer the hand gets to a good grasp position, the more it blocks the camera's view of exactly what it's about to grab. Payload-to-hand-weight ratio varies a lot across current dexterous hands, and it's a meaningful tradeoff - more DoF and finer sensing generally means more actuators and mass in the hand itself, which eats into the arm's usable payload budget.

Tangent, but worth mentioning: Object occlusion by the robot's own hand during the final approach to a grasp is a common and annoying perception problem - the closer the hand gets to a good grasp position, the more it blocks the camera's view of exactly what it's about to grab.

Not sure I fully agree here. Vision-based grasp confidence estimation (predicting success before attempting a grasp) and tactile-based confirmation (confirming after contact) are complementary rather than competing - vision helps you choose a grasp, tactile tells you if it actually worked. Wet, oily, or otherwise low-friction objects are still a genuine edge case for most current hands, since tactile sensing and grasp-force controllers are typically tuned and validated on dry, higher-friction test objects.

New to this, so forgive me if this is obvious - Cable routing through a wrist joint with multiple degrees of freedom is a genuinely tricky mechanical design problem - tendons and wiring both need enough slack to avoid binding through the full range of motion without tangling or fraying over thousands of cycles. Reminds me a bit of the early drone hobbyist scene, honestly.

@ethan_fisc Thanks for laying this out, genuinely useful. Imitation learning from human demonstration video (without robot teleoperation data) is an appealing way to scale up training data cheaply, but it runs into the embodiment gap - human hand kinematics and force profiles don't map directly onto a robot hand's very different mechanism. Underactuated hands (fewer actuators than joints, using mechanical coupling to shape the grasp) are a reasonable engineering compromise for robust power grasps on a budget, but they generally can't do fine in-hand manipulation the way a fully actuated hand can.

This is exactly the kind of context I was looking for. Payload-to-hand-weight ratio varies a lot across current dexterous hands, and it's a meaningful tradeoff - more DoF and finer sensing generally means more actuators and mass in the hand itself, which eats into the arm's usable payload budget.

@jonathan.rao1 Not to derail, but this reminds me of something adjacent: Payload-to-hand-weight ratio varies a lot across current dexterous hands, and it's a meaningful tradeoff - more DoF and finer sensing generally means more actuators and mass in the hand itself, which eats into the arm's usable payload budget. This whole thread is a good reminder how young this field still is.

This matches something I went through recently. Grasp planning for deformable or non-rigid objects (bags, cables, cloth) remains one of the genuinely unsolved problems in manipulation - rigid-body grasp models simply don't capture how the object will behave once contact starts. Vision-based grasp confidence estimation (predicting success before attempting a grasp) and tactile-based confirmation (confirming after contact) are complementary rather than competing - vision helps you choose a grasp, tactile tells you if it actually worked. Makes me wonder how this looks in another five years.

@kwilliams Not to derail, but this reminds me of something adjacent: In-hand reorientation (repositioning a grasped object without setting it down) is one of the more advanced manipulation skills, requiring either a highly dexterous hand with enough DoF or clever use of gravity and controlled slipping - it's an active research area rather than a solved problem. Vision-based grasp confidence estimation (predicting success before attempting a grasp) and tactile-based confirmation (confirming after contact) are complementary rather than competing - vision helps you choose a grasp, tactile tells you if it actually worked.

This is a great summary, thanks. Wet, oily, or otherwise low-friction objects are still a genuine edge case for most current hands, since tactile sensing and grasp-force controllers are typically tuned and validated on dry, higher-friction test objects. Vision-based grasp confidence estimation (predicting success before attempting a grasp) and tactile-based confirmation (confirming after contact) are complementary rather than competing - vision helps you choose a grasp, tactile tells you if it actually worked.