Simple Torque-Observation Alignment for Zero-Shot Sim-to-Real Grasping with a Direct-Drive Gripper

Doyoung Kim; Edgar Lee; Hyeonsun Park; Chunghyeon Lee; Chihyun Han; Uisu Hwang; Seokhwan Jeong

cs.ROSep 24, 2026

Simple Torque-Observation Alignment for Zero-Shot Sim-to-Real Grasping with a Direct-Drive Gripper

Authors: Doyoung Kim, Edgar Lee, Hyeonsun Park, Chunghyeon Lee, Chihyun Han, Uisu Hwang, Seokhwan Jeong

Organizations: Department of Mechanical Engineering, Sogang University, Seoul, South Korea

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.

Figures & tables

Fig. 2: Simple torque observation alignment method pipeline.

Fig. 3: Direct-drive gripper platform. (a) DD actuator-based multifingered gripper. (b) Direct-Drive (DD) actuators integrated into a single finger. (c) Joint and link configuration.

Fig. 4: Dynamometer-based identification of the effective torque constants. (a) Experimental setup consisting of a test motor, a load motor, and an inline torque sensor. (b) Measured shaft torque versus quadrature current for the GL40 motor. (c) The corresponding result for the GL60 motor. Dashed lines denote the linear regression fits.

Motor	Datasheet $K_{\tau}$	Calibrated $K_{\tau}^{\ast}$	$R^{2}$	RMSE [Nm]
Motor	[Nm/A]	[Nm/A]	$R^{2}$	RMSE [Nm]
GL60 KV25	0.4500	0.3260	0.9940	0.0098
GL40 KV70	0.1500	0.1392	0.9992	0.0023

TABLE I: Datasheet and dynamometer-identified torque constants with regression accuracy.

Motor	Gaussian model $\epsilon_{m,t}$	Injected std. $\sigma_{m}$ [Nm]
GL60 KV25	$\mathcal{N}(0,9.688465{\times}10^{-5})$	0.009843
GL40 KV70	$\mathcal{N}(0,5.313025{\times}10^{-6})$	0.002305

TABLE II: Motor-type-specific Gaussian noise derived from dynamometer regression residuals and injected into simulated torque observations.

Fig. 5: Teacher-student policy training and deployment pipeline.

Teacher (PPO)
Parameter	Value
Learning rate	$10^{-3}$
Steps per environment	24
PPO update epochs	5
Minibatches	6
Activation	ELU

TABLE III: Training hyperparameters.

Shape	Dimensions [cm]	A	B	C
Cuboid	Base length $\times$ height	$6\times 6$	$8\times 8$	$10\times 8$
Cylinder	Radius $\times$ height	$4\times 4$	$4.5\times 4.5$	$5\times 5$
Sphere	Radius	$4$	$4.5$	$5$

TABLE IV: Dimensions of the training objects.

Parameter	Range
Robot: Initial MCP Joint Offset (rad)	$+\mathcal{U}[-0.06,\,0.06]$
Robot: Initial PIP Joint Offset (rad)	$+\mathcal{U}[-0.35,\,0.35]$
Object: Initial Position ( $x$ , $y$ ) (m)	$+\mathcal{U}[-0.01,\,0.01]$
Object: Initial Yaw (rad)	$\mathcal{U}[-\pi,\,\pi]$
Object: Mass	$\times\mathcal{U}[0.5,\,3.0]$
Actuator: P (Stiffness) Gain	$\times\mathcal{U}[0.85,\,1.05]$

TABLE V: Randomization applied during teacher training.

Fig. 7: Objects used in the real-world evaluation. The evaluation includes 21 objects: nine in-distribution (ID) objects and twelve out-of-distribution (OOD) objects. The ID objects belong to the cuboid, cylinder, and sphere categories used during simulation training. The OOD objects are real-world items with previously unseen shapes, materials, and surface properties that fit within the gripper workspace.

Policy	Observation	Noise	Success rate
Policy	Observation	Noise	Simulation	Real-world
Position Only	$[q_{t}]$	N/A	68.9%	15.6%
Scale + Noise	$[q_{t},\tau_{t}]$	On	100.0%	26.7%
Scale + Bias	$[q_{t},\Delta\tau_{t}]$	Off	100.0%	62.2%
Our Method	$[q_{t},\Delta\tau_{t}]$	On	100.0%	100.0%

TABLE VI: Simulation and real-world success rates of student policies

\pi_{s}

ID Objects			OOD Objects
#	Object	Rate	#	Object	Rate
1	Cuboid A	100%	10	Box	100%
2	Cuboid B	100%	11	Stuffed toy	100%
3	Cuboid C	100%	12	Tennis ball	90%
4	Cylinder A	100%	13	Plastic container	100%
5	Cylinder B	100%	14	Wine glass	100%

TABLE VII: Real-world grasping success rates on ID and OOD objects.

Fig. 8: Representative real-world grasping sequences and motor-current-derived $\Delta\tau$ trajectories for the complete method and two $\Delta\tau$ ablations on the DD gripper. The sequences show grasp outcomes alongside the torque-change observations. (a) Our Method , using calibrated $K_{\tau}^{\ast}$ and noise injection, completes the grasp–lift–hold task. (b) Scale + Bias , using calibrated $K_{\tau}^{\ast}$ without noise injection, exhibits a grasping/lifting failure. (c) Bias + Noise , using the datasheet $K_{\tau}$ with noise injection, exhibits a holding failure. In the plot legends, F1–F3 denote the three fingers, while MCP, PIP, and DIP denote the three joints of each finger.

Explore similar work

Closing the Reality Gap: Zero-Shot Sim-to-Real Deployment for Dexterous Force-Based Grasping and Manipulation
Jul 6, 2026Zhe Zhao, Zhibin Li, Yilin Ou +1Contact-Rich Dexterous Manipulation Sim-To-Real Reinforcement Learning
State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications, China · University College London, United Kingdom
TAM: Torque Adaptation Module for Robust Motion Transfer in Manipulation
Jun 4, 2026Dongwon Son, Florian Shkurti, Jason Lee +3Torque Robotic Manipulation Policies
1KAIST · *Work done during an internship at the Allen Institute for AI. · 2Allen Institute for AI +2
Actuator Reality Shaping for Zero-Shot Sim-to-Real Robot Learning
Jul 2, 2026Satoshi Yamamori, Koji Ishihara, Kenjiro Minamikawa +4Sim-To-Real Reinforcement Learning Scalable Robot Learning
Graduate School of Informatics, Kyoto University, Kyoto, Japan · Dept. of Brain Robot Interface, Computational Neuroscience Labs, ATR, Kyoto, Japan

Jul 6, 2026, cs.ROJ/K move · Enter open · S save

Closing the Reality Gap: Zero-Shot Sim-to-Real Deployment for Dexterous Force-Based Grasping and Manipulation

Zhe Zhao, Zhibin Li, Yilin Ou, Mengshi Qi

State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications, China · University College London, United Kingdom

Contact-Rich Dexterous Manipulation Sim-To-Real Reinforcement Learning Dexterous Manipulation Bimanual Manipulation Tactile Torque+1

Open paper AskPDF arXiv

Human-like dexterous hands with multiple fingers offer human-level manipulation capabilities but remain difficult to train the control policies that can deploy on real hardware due to contact-rich physics and imperfect actuation. We present a sim-to-real reinforcement learning method that leverages dense tactile feedback combined with joint torque sensing to explicitly regulate physical interactions. To enable effective sim-to-real transfer, we introduce (i) a computationally fast tactile simulation that computes distances between dense virtual tactile units and the object via parallel forward kinematics, providing high-rate, high-resolution touch signals needed by RL; (ii) a current-to-torque calibration that eliminates the need for torque sensors on dexterous hands by mapping motor current to joint torque; and (iii) actuator dynamics modeling with randomization to account for non-ideal torque-speed effects and bridge the actuation gaps. Using an asymmetric actor-critic PPO pipeline, we train policies entirely in simulation and deploy them directly to a five-finger hand. The resulting policies demonstrate two essential human-hand skills: (1) command-based controllable grasp force tracking and (2) reorientation of objects in the hand, both of which are robustly executed without fine-tuning on the robot. By combining tactile and torque in the observation space with scalable sensing and actuation modeling, our system provides a practical solution to achieve reliable dexterous manipulation. To our knowledge, this is the first demonstration of controllable grasping on a multi-finger dexterous hand trained entirely in simulation and transferred zero-shot on real hardware.

Abstract

Explore similar work

Closing the Reality Gap: Zero-Shot Sim-to-Real Deployment for Dexterous Force-Based Grasping and Manipulation

TAM: Torque Adaptation Module for Robust Motion Transfer in Manipulation

Actuator Reality Shaping for Zero-Shot Sim-to-Real Robot Learning