cs.ROSep 24, 2026

Real-Time Force Regulation for Whole-Hand Dexterous Grasping

Authors: Sang Min Kim, Alexander Alexiev, Tzu-Yuan Lin, Sangbae Kim, Young Min Kim, Yonghyeon Lee

Organizations: Department of Electrical and Computer Engineering, Seoul National University, Seoul, Republic of Korea. · Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA. · Department of Artificial Intelligence, Yonsei University, Seoul, Republic of Korea.

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Robust dexterous grasping requires maintaining physical stability despite contacts interactively evolving across the entire hand. A precomputed force distribution can easily fail under object motion, modeling errors, or external disturbances. In this paper, we present a framework for real-time force regulation over dynamically changing whole-hand contacts. Our method geometrically estimates contacts across all hand links using a tracked object model and proprioception, without requiring tactile sensing at those contacts. It repeatedly recomputes the desired contact-force distribution subject to friction constraints, actuator limits, and an actuation-consistency constraint motivated by classical whole-limb force analysis. We integrate this force-regulation controller with reactive reaching, enabling the hand to acquire a grasp, maintain it under disturbances, and regrasp after losing the object. Simulation experiments without gravity demonstrate improved grasp retention over fixed-allocation and fingertip-only execution under controlled perturbations, while real-world experiments on a 27-DoF arm-hand system demonstrate grasp maintenance and recovery under human-applied disturbances as contacts evolve across the whole hand. Project page: https://sangminkim-99.github.io/reactive-grasp-whole-hand/

Figures & tables

Fig. 2: Visualization of perturbation-test protocol. Starting from a stored grasp, the controller regulates the contact forces (a, b), and a force equal to 10% of the object weight is applied along one of ±x,±y,±z\pm x,\pm y,\pm z at a time (c). A spring–damper wrench then pulls the object toward a pose rotated by θ\theta about a random axis (d; grey/red dashed outline: before/after), and the load test is repeated (e). A trial succeeds only if the object is held through all six directions.
Fig. 5: Effect of the actuation-consistency constraint on a three-finger grasp, 2 s after initialization. Blue: desired contact forces; green: measured contact forces; red: measured net contact force at the COM, nearly zero with the constraint. Blue arrows use twice the length scale of green arrows.
Fig. 9: Execution snapshots of the full pipeline on three objects, combining hierarchical reactive reaching [ 20 ] with the whole-hand force regulation. Frames with human perturbation are labeled in orange. The hand follows the object displaced during reaching, regulates the grasp under disturbances after contact, and, when the object slips or is snatched away, returns to reaching and regrasps it at its new pose. Timestamps are in seconds.
ObjectDuration [s]PerturbationsLossesRecovered
Pringles591633/3
Small box483121/2
Large box943132/3
TABLE I: Summary of full-pipeline runs across three objects. Losses denote the hand losing the object, while unrecovered losses correspond to workspace exits.
Fig. 10: A manually posed grasp with fingers alternating in front of and behind the object, regulated by the whole-hand controller. Bottom: palmar and dorsal views colored by distance to the nearest contact. Cyan ellipses show the change in estimated contact on the little finger’s dorsal surface.

Explore similar work

CardsList

  1. Interaction Dynamics for Dexterous Manipulation

    Jun 12, 2026Yongyan CaoDexterous Robot HandsDexterous Manipulation

  2. Transferring Contact, Not Just Motion: Compliant Grasping Across Dexterous Hands

    Jun 14, 2026Soofiyan Atar, Yao-Ting Huang, Michael YipDexterous Robot HandsCross-Embodiment Transfer

  3. CoorGrasp: Coordinated Contact Control for Adaptive Dexterous Grasping Under Uncertainty

    Jul 3, 2026Mingrui Yu, Yongpeng Jiang, Yongyi Jia +2Grasp GenerationHand