Torque alignment improves zero-shot simulated robotic grasping success
Simple Torque-Observation Alignment for Zero-Shot Sim-to-Real Grasping with a Direct-Drive Gripper
Robotics
Summary
Robots need to feel and react to forces when they pick things up, but the way these forces are measured in simulations can differ from the real world. The researchers found a simple way to adjust the force measurements to match better between simulation and real robots with direct-drive grips. They trained a robot hand in simulation and used their method to make it work perfectly when picking up real objects without extra training. This method makes it easier to use computer-trained skills directly on real robots.
What this means in practice
- •For robotics engineers: Improve real-world grasping performance of simulated-trained direct-drive robotic grippers without additional real-world tuning.
- •For automation system designers: Deploy zero-shot grasping policies on robotic hands using torque difference measurements to handle varied objects reliably.
Authors
Doyoung Kim, Edgar Lee, Hyeonsun Park, Chunghyeon Lee, Chihyun Han, Uisu Hwang, Seokhwan Jeong
Abstract
Torque observations in reinforcement learning remain challenging because simulated and measured torque differ in scale, offset, and noise. In this paper, we propose a simple torque observation alignment method for robots with direct-drive (DD) actuators, in which motor current maps linearly to joint torque through a motor-type-specific torque constant K_tau. First, dynamometer calibration identifies K_tau* and corrects the scale mismatch between simulated and real torque. Second, the method uses delta_tau(t) = tau(t) - tau(t-1) as the observation in both domains to eliminate the constant offset instead of using the direct torque tau(t), which carries a domain-dependent bias. Third, Gaussian noise obtained from the dynamometer measurement data is injected during the learning process. To validate the proposed method, we train a teacher-student grasping policy entirely in simulation and deploy the distilled student on a multifingered DD gripper. The deployed policy performs proprioceptive grasping using only joint positions and torque differences. We conduct an ablation study comparing the proposed method with alternative alignment variants on nine in-distribution (ID) objects. The proposed method achieves 100% grasp success. These results demonstrate that the proposed alignment method improves the robustness of zero-shot policy transfer on the DD gripper against real-world torque-observation mismatches.