Atomic motion coordinate improves language-steered robot arm actions
Atomic Motion Coordinate for Language-Steerable and Force-Responsive Manipulation
Robotics
Summary
This paper looks at how robot arms can better follow language instructions instead of just relying on visual cues. The researchers introduce a new way to represent movements called Atomic Motion Coordinate, which breaks down arm motions into smaller parts tied to text and robot balance. This method helps robots change their actions based on language and the forces they feel. The authors tested this approach offline and in real robots, showing better success in manipulating objects and adapting to forces than previous methods.
What this means in practice
- •For industrial robot integrators: Enable robot arms to better follow spoken or written instructions for diverse tasks by using atomic motion coordinates to adapt motions without heavy reliance on vision.
- •For robotic grasping engineers: Improve robotic manipulation under uncertain contact forces by incorporating force-responsive atomic motion adjustments for more reliable grasping and placement.
Authors
Jiaqi Zhai, Jingkai Zhao, Chen Yang, Siyuan Ma, Yutian Zhang, Liwen Yang, Qinglian Wu, Weiqi Fan, Yifei Wang, Yi Zheng, Chenxi Gu, Dong Wei, Wei Zhang
Abstract
Can changing only the language instruction redirect a VLA policy's end effector, or does the visually driven motion prior dominate? We present Atomic Motion Coordinate, a geometry-grounded coordinate for steerable and force-responsive manipulation. Each arm owns thirteen signed translation, rotation, and hold atoms grounded from text and forward kinematics with vision withheld, and the coordinate is injected into every action-expert block via weighted codebook alignment. Contact history modulates the same coordinate through a bounded spherical residual that is recomputed from a fixed nominal latent to regenerate only the unexecuted horizon suffix. Across 7,520 offline horizon interventions, opposite-atom separation reaches 92.5/83.1% (single/dual) versus 39.1/24.0% for LA4VLA-style. Across 50 real-robot trials per task, AMC raises OOD fruit progress from 60.5% to 87.8%; force adaptation raises Plug/Vase from 59.0/71.5% to 78.5/75.2%.