Tactile feedback improves fine control of robot grippers during tasks
Touch Where It Matters: Learning Local Tactile Policies for Fine-Grained Gripper Control
Robotics
Summary
Robot arms often use cameras to guide their movements, but handling objects precisely, especially when touching them, needs more detailed control. The authors created a system where a robot's gripper uses touch and force sensors to adjust its hold during delicate parts of a task, while the arm keeps moving under vision control. This approach helps the robot open containers, handle objects, and use tools better than previous methods that relied only on vision or simpler touch feedback. The study shows that combining touch information directly into local control can make robotic grasping more precise and adaptable.
What this means in practice
- •For robotics engineers: Control robot grippers with tactile feedback to improve precision during complex object manipulation tasks.
- •For automation system integrators: Enhance industrial robot handling by adding localized touch-based gripper control alongside vision-guided arm movement.
Authors
Mingyuan Ba, Yanan Zhou, James Zhao, Weiming Zhi
Abstract
Visual imitation policies can organise task motion, yet contact-rich object use demands gripper adjustments beyond coarse opening and closing. We study fine-grained gripper control as a local feedback problem within ongoing task execution. Our approach assigns tactile feedback a dedicated control role during contact-rich task stages. We introduce ContactPolicy, a learned local controller that takes over the gripper while a visual base policy continues to control the arm. ContactPolicy combines tactile and force histories with current robot state to jointly predict gripper increments and their modulation, updating the gripper target from each new observation. An Intent policy manages control handover between the visual base policy and ContactPolicy. Real-world experiments on container opening, object handling, and tool operation show improved task success over both the Visuomotor Policy and Tactile-Force Policy baselines. Execution analysis illustrates how the gripper adjusts contact within an ongoing task, including periods of target retention and fine-grained changes as interaction conditions evolve. Arm-state ablations and failure analysis further examine how execution context supports local regulation. These results support task-local tactile policies for fine-grained gripper control within visual manipulation.