What's your best tip for debugging a grasp policy that fails inconsistently?

[ivan22]

This came up in a Discord I'm in and I wanted a more permanent place to discuss it. Palm sensing gets less attention than fingertip sensing, but a lot of power grasps (holding a box, a tool handle) rely more on palm and lateral finger contact than fingertip contact, so under-sensing the palm can leave a real blind spot in grasp confidence. 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. Anyone want to poke holes in this?

[servoken70]

@ivan22 Counterpoint: 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.

[mohammed64]

Not sure I fully agree here. 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.

[kwameivanov]

@mohammed64 Appreciate the detailed answer. Compliant wrists that absorb impact during a bad approach or misjudged contact reduce mechanical stress on the whole arm, which matters a lot for long-term reliability even though it's a less visible feature than the hand itself.

[edward.nelson]

Can I ask a dumb follow-up - 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.

[timothy.roberts2]

@edward.nelson Pretty much this. One thing to add: Figure's fourth-generation Dexterous Hand (on Figure 02/03) reportedly offers 16 degrees of freedom per hand with sensors integrated into each finger, aimed at fine force control tasks like handling small electronic components without crushing them. 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.

[carol38]

@timothy.roberts2 Small correction on one detail: 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.

[barbara.jones]

Side note that might be relevant: Bimanual manipulation - two arms coordinating on one task - is harder than it looks mostly because of the added degrees of freedom and the timing/force coordination required; a lot of 'two-handed' demos are actually closer to two independent single-hand tasks done in sequence.

[amara.brown]

That's the official framing, at least - reality tends to lag a bit. A lot of the manipulation shown in production demos still leans heavily on teleoperation, particularly for anything involving fine force control or novel objects - autonomous grasp success rates on genuinely unstructured, previously-unseen clutter are still well below what teleoperation can achieve.

[priya85]

Same conclusion I've come to. Also worth noting: 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.