Robot learns gentle grasping from tactile reflex demonstrations
What is the Better Curriculum: Controller-Shaped Grasping Behavior for Contact Force-Sensitive Manipulation
Robotics
Summary
Grasping fragile objects without breaking them is very hard for robots because they need to feel and control very light touch forces. The authors show that instead of teaching a robot a complex policy using touch data, it is better to use a simple fast reflex controller to collect demonstration data. Then, the robot learns from these demonstrations without using touch directly at run time. This approach helps the robot reliably hold objects like plastic cups, but still needs the reflex controller to handle surprises.
What this means in practice
- •For robot developers: Train robot grasping policies using tactile-reflex shaped demonstrations to improve stable grasping of delicate objects.
- •For automation engineers: Enhance fragile-object handling systems by combining learned policies with reflex controllers for disturbance rejection.
Authors
Ziyan Feng, Zizhao Yuan, Yulong Fu, Yuxin He, Zhiyuan Zhang, Zhengjie Zhang, Jinni Zhou, Renjing Xu, Qiang Nie
Abstract
How should a robot learn to manipulate objects so fragile that sub-Newton contact forces can cause irreversible damage? Existing visuo-tactile policy learning typically treats tactile sensing as an additional policy input. In direct-contact force-sensitive manipulation, however, the bottleneck can arise earlier, during data collection: manual gripper control is too delayed and coarse-grained to reliably maintain the narrow force range required for stable grasping. We therefore use a deterministic 25 Hz tactile reflex controller as a collection-time teacher, producing demonstrations with controller-shaped grasping behavior for tactile-free policy learning. On Action Chunking with Transformers (ACT), policies trained from reflex-shaped demonstrations recover the teacher's grasping profile and achieve 95% stable grasps on the nominal plastic-cup task, substantially outperforming visually screened manual demonstrations. The same intervention improves in-distribution stability on $π_{0.5}$ and shows a favorable exploratory trend on an unseen paper-cup variant. Under randomized external disturbance, however, the reflex-data $π_{0.5}$ policy still fails in 45% of policy-only trials, whereas a deployment-time reflex arbiter retains all grasps. These results reveal a new role for tactile feedback in force-sensitive manipulation: rather than integrating tactile into the policy, we use it as a collection-time teacher that shapes grasping behavior in demonstrations for policy learning, while disturbance rejection remains controller-dependent, revealing the boundary of tactile-free policy.