Structured model improves robot force-sensitive manipulation accuracy
CW-WAM: A Structured Contact-Wrench World Action Model for Force-Sensitive Manipulation
Robotics
Summary
Robots sometimes have trouble feeling how much force they're applying when moving or touching things, which makes delicate tasks hard. The authors created a new model called CW-WAM that helps robots predict both motion and the forces they feel during manipulation. This makes robots better at tasks like inserting objects carefully or lifting things with different weights by improving force feedback and control. Their system greatly increases success rates in tricky insertions and reduces force errors compared to previous methods.
What this means in practice
- •For robotics engineers: Improve robotic insertion and lifting tasks by incorporating force-aware predictions to increase precision and success rates in challenging contact scenarios.
- •For industrial automation teams: Enhance force feedback control in automated assembly systems for delicate or tight-tolerance parts to reduce errors and improve process reliability.
Authors
Houzhu Ding, Xinda Qi, Luqi Wang, Ming Yang
Abstract
Force-sensitive manipulation requires contact and load information that is not provided explicitly by visual observations. We present CW-WAM, a world action model that jointly predicts video, motion, and contact wrench. Inspired by hybrid motion-force control, it contains a unified interaction module that treats both motion and contact wrench as modeling targets and predicts time-aligned motion-wrench chunks as control references. A pair-causal attention mechanism structures the motion-wrench dependencies, while a long fixed-length wrench memory retains past interaction information. The predicted wrench chunk acts as a force reference for a 20-Hz outer-loop admittance controller within a slow-fast control system that maintains force feedback and smooth motion between model queries. CW-WAM improves contact prediction and increases the insertion success rate from 25% to 70% for square pegs with 1-mm clearance per side and from 40% to 95% for power adapters, compared with WAM without wrench input. It also improves lifting-height accuracy across different loads. Our framework with an admittance controller reduces force-reference RMSE from 4.8 to 2.7 N for peg insertion and from 2.0 to 1.3 N for wiping, respectively, with smaller force fluctuations during successful execution. Project website: https://cw-wam-project.github.io/