DexForce

Extracting Force-informed Actions from Kinesthetic Demonstrations

for Dexterous Manipulation

Claire Chen, Zhongchun Yu, Hojung Choi, Mark Cutkosky, Jeannette Bohg

Stanford University

RA-L June 2025

Read the paper Watch a 5 minute summary (captioned) BibTex

DexForce leverages contact forces, measured during kinesthetic demonstrations, to compute force-informed actions.

We use DexForce demonstrations to learn policies for6 tasksand evaluate each rollout according to the following criteria:

slide cube

success

failure

reorient smiley

success

partial success

failure

open AirPods

success

partial success

failure

grasp battery

success

partial success

failure

unscrew nut

success

partial success

failure

flip box

success

partial success

failure

We show that DexForce's force-informed actions are critical for learning capable policies.

Policies trained on our force-informed actions achieve an average76%success rate across all 6 tasks.
Baseline policies trained on actions thatdo notaccount for contact forces havenear zerosuccess rates.

To view policy rollouts trainedwithandwithoutforce-informed actions for each task, view on larger screen.

Explore policies trainedwithandwithoutforce-informed actions for each task:

with force-informed actions

without force-informed actions

with force-informed actions

without force-informed actions

with force-informed actions

without force-informed actions

with force-informed actions

without force-informed actions

with force-informed actions

without force-informed actions

with force-informed actions

without force-informed actions

Hardware

We instrument an Allegro hand with a wrist-mounted Intel RealSense D435 camera (+ fisheye lens attachment) and two CoinFT six-axis force-torque sensors.

To re-create our setup:
Fisheye RealSense: We have open-sourced our camera and fisheye lens mounting parts. Get the instructions and partshereand pleasecite usif you find these useful!
CoinFT: The developers of CoinFT plan to release the sensor design soon!

Supplementary video


                @ARTICLE{chen2025dexforce,
                    author={Chen, Claire and Yu, Zhongchun and Choi, Hojung and Cutkosky, Mark and Bohg, Jeannette},
                    journal={IEEE Robotics and Automation Letters}, 
                    title={DexForce: Extracting Force-Informed Actions From Kinesthetic Demonstrations for Dexterous Manipulation}, 
                    year={2025},
                    volume={10},
                    number={6},
                    pages={6416-6423},
                    keywords={Robots;Force;Robot kinematics;Robot sensing systems;Hands;Force measurement;Sensors;Impedance;Imitation learning;Position measurement;Dexterous manipulation;force and tactile sensing;imitation learning},
                    doi={10.1109/LRA.2025.3568318}
                }