cs.ROSep 12, 2026

GIFT: Glove-Inferred Force Transfer: Force-Aware Human-to-Robot Skill Transfer from a Wearable Sensing Glove to a Robot Hand Without Tactile Sensors

Authors: Tzach Sarusi

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Human-to-robot skill transfer from sensing gloves has so far relied on shared hardware: the same tactile glove worn by the demonstrator and the robot, or a learned alignment between two tactile sensors. We present GIFT (Glove-Inferred Force Transfer), a pipeline in which the interface between human and robot is a physical unit rather than a shared sensor: fingertip force is measured in newtons on the human side and estimated in newtons on the robot side. A wearable glove records finger flexion, calibrated fingertip force, and wrist orientation, while a head-mounted camera records the demonstration; no robot is present. At deployment, the robot estimates force from actuator-current residuals relative to a free-space baseline, through a calibrated mapping to newtons, so any position-controlled hand that reports motor current can serve as the deployment platform. The policy uses a glove-space state and predicts finger-position targets; the robot enters only through two calibrated adapters, a retargeting decoder and a force estimator. We evaluate GIFT on a cup grasp-and-hold task with two action-chunking policies trained on the same demonstrations, with fingertip-force inputs retained in one and zeroed in the other. In a 50-rollout evaluation with sample size and metrics fixed before scoring, both policies succeeded in all 25 rollouts. The median of the per-rollout hold-phase grip-force estimates was 53% lower with force inputs: 1.20 N versus 2.55 N (one-sided Mann-Whitney U, p<0.0001). In an observation ablation, a vision-only policy achieved 0/15 grasps, policies given hand-command state acquired the grasp, and the force inputs determined how hard the policy held. A force channel measured on the human hand thus transfers to a robot hand with no tactile hardware, through a retargeting map from five glove channels to seven robot actuators, with no sensor shared between the two.

Figures & tables

Fig. 1: GIFT. Top: the sensing glove (fingertip sensors highlighted) captures fingertip force in newtons, finger curl and wrist orientation, alongside a head camera; no robot is present during collection. Bottom: the policy trained on these demonstrations is deployed on a robot hand with no tactile sensors, whose force observation is estimated in newtons from its own actuator currents. The human and the robot share a physical unit, not a sensor.
Fig. 2: Hardware. (A, B) The sensing glove, palm and dorsal sides, with sensor locations highlighted: five fingertip force-sensitive resistors (discs) and five flex sensors along the fingers (streaks); the wrist enclosure holds the ESP32-S3, the IMU and the LED/buzzer start cue. (C) Deployment setup (illustration): the TetherIA Aero Hand Open in a plain nitrile glove with no sensors, mounted on an SO-101 arm, the cup at its marked position, and the camera on a fixed mount aligned to the demonstrator’s viewpoint. (D) Capture signal path: all glove channels and the camera share one time axis on the recording host.
Fig. 3: Two instruments, one unit. (A) Glove side: fingertip FSR readings against a digital reference scale; the shaded band is the hold-phase grip range reported in § IV-D . (B, C) Robot side: actuator current against commanded position for the thumb-tendon and index actuators, from the free-space log (no object) with its fitted linear baseline, and from a contact log. At the grasp position the contact current rises vertically above the baseline; that residual, mapped to newtons, is the policy’s deployment-time force observation.
Fig. 5: Representative rollouts from the head camera. Top row: a human demonstration with the sensing glove. Middle rows: the with-force and without-force policies on the robot hand, showing for each arm the run whose hold-phase median is closest to its arm’s median (1.20 N and 2.55 N). Bottom row: the vision-only policy lands the thumb on the cup, bottoms out the close without a grasp, and the arm lifts an empty hand. Right: the logged grip-force estimate over the two state-policy runs shown, with their hold-phase medians dashed.
ObservationsGraspHoldGrip force (N)
Vision only0/15——
+ hand state (no touch)25/2525/252.55
+ fingertip force25/2525/251.20
TABLE I: Observation ablation on the cup grasp-and-hold task. Hand-command state acquires the grasp; the fingertip-force channel changes how hard the policy holds, not whether it holds. Grip force is the median of per-rollout hold-phase medians ( n=25n=25 per policy). All completed rollouts reported.

Explore similar work

CardsList
May 6, 2026cs.RO

Active Contact Sensing for Robust Robot-to-Human Object Handover

Robot-to-human object handover is an essential skill for robot assistants, from serving drinks at home to passing surgical tools in the operating room. We expect robots to perform handover robustly -- to release the object only after a firm human grasp while ignoring incidental touches. Existing passive-sensing methods struggle to generalize across diverse objects and human behaviors, as they lack informative perturbations to disambiguate different contact conditions, such as firm grasp versus incidental touch. We propose an active sensing approach for robust handovers: the robot applies information-gathering motions and senses the resulting human-applied forces to infer the contact state. A firm grasp produces forces in multiple directions, while an accidental touch does not. To capture this distinction, we model the contact state with a Bayesian linear model: a distribution over piecewise-linear mappings from robot motions to human-applied forces. This model enables firm grasp detection and active information gathering. In experiments with 12 participants and 30 diverse rigid objects, our method achieved a 97.5% success rate -- over 30% higher than two common baselines.

Linfeng Li, Lin Shao, David Hsu

Robotic HandsHandoffHuman AgencyObjects
AskPDF
Jul 31, 2026cs.RO

TacPrint: A Wearable Fingertip Tactile Sensor for Human-to-Robot Contact Reproduction

Human-centric data collection is emerging as a significant paradigm for robot skill acquisition, but seamlessly integrating low-cost, scalable tactile sensing systems that capture fine-grained fingertip interactions without compromising natural operation remains a key challenge. This reduces the reliability of human-to-robot transfer in contact-rich tasks. In this work, we present TacPrint, a wearable fingertip tactile sensor, where protrusions on the inner surface of the silicone skin are aligned one-to-one with 24 capacitive taxels to enable localized capacitive responses. A real-to-sim-to-real pipeline estimates a 35 ×\times 26 contact-depth map from 24-channel capacitive signals. Against simulation-generated labels, the model achieved a contact-region RMSE of 0.223 ±\pm 0.161 mm, a weighted-centroid error of 1.213 ±\pm 2.379 pixels, and an IoU of 0.829 ±\pm 0.169. With measured capacitive inputs, the network-predicted depth evaluated at the guide-calibrated contact center showed a mean absolute error of 0.085 ±\pm 0.057 mm across all 40 controlled trials, while the mean contact-position error was 0.250 ±\pm 0.208 mm across the 37 trials whose reference contact regions were not truncated by the sensing boundary. In human-to-robot replay, tactile-guided compensation increased grasping and wiping success rates from 0% to 91.67% and 90%, respectively. In closed-loop grasping, dense-depth feedback achieved success rates of 87.5% over all tested positions and 85% under edge-contact conditions, compared with 67.5% and 45% for raw-taxel feedback.

Yongxi Liu, Chaofan Zhang, Xingyu Zhang +4

Institute of Automation, Chinese Academy of Sciences · Imprintx Robotics · Beijing Academy of Artificial Intelligence

TactileHuman-To-Robot TransferHuman-Robot InteractionSensorsRegion
AskPDF
Jun 14, 2026cs.RO

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

Dexterous grasping depends on contact regulation, not motion alone. Stable manipulation requires fingers to maintain appropriate object loading as contacts slip, deform, or become visually occluded. Existing cross-embodiment dexterous policies unify motion through retargeted hand poses or latent actions, but force feedback remains tied to each hand's sensing and actuation, limiting transfer. This work introduces a cross-embodiment force-position interface for contact-aware manipulation across heterogeneous dexterous hands. Motion intent is represented in a shared hand-pose latent, while each hand's effort signal is calibrated through system identification into physical joint torque in N.m. These torques are mapped to fingertip forces and compact per-finger load descriptors, giving the policy comparable observations of where the hand should move and how the object is loaded. Using this interface, a flow-matching visuomotor policy is trained on vision, proprioception, and calibrated contact, with structured visual masking that encourages reliance on force under grasp-relevant occlusion. The same calibrated signal drives a hybrid force-position controller for demonstration collection and execution, keeping force targets consistent across training and deployment. Experiments across structurally different hands show that calibrated contact feedback enables transferable compliant grasping, with learned primitives reusable in long-horizon manipulation pipelines.

Soofiyan Atar, Yao-Ting Huang, Michael Yip

Department of Electrical and Computer Engineering University of California San Diego, United States

Dexterous ManipulationBimanual ManipulationCross-Embodiment TransferVisuomotor PolicyContactMotion+1
AskPDF