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

2026-01-06

arXiv.org

AI Summary

This work addresses the challenge of deploying control policies for multi-fingered dexterous hands in real-world settings, where complex contact dynamics and non-ideal actuator characteristics hinder performance. The authors propose a reinforcement learning–based zero-shot sim-to-real transfer framework that integrates dense tactile sensing and joint torque feedback. By employing high-fidelity simulation—including fast tactile rendering, current-to-torque calibration, actuator dynamics, and randomized modeling of non-ideal effects such as backlash and torque–velocity saturation—the policy is trained entirely in simulation and deployed directly on a physical five-fingered dexterous hand without fine-tuning. This approach achieves, for the first time, precise grasp-force tracking and in-hand object reorientation using a purely simulation-trained policy, demonstrating both efficacy and practicality.

Technology Category

• Intelligent Robots: Manipulation
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments

Abstract

Human-like dexterous hands with multiple fingers offer human-level manipulation capabilities, but training control policies that can directly deploy on real hardware remains difficult due to contact-rich physics and imperfect actuation. We close this gap with a practical sim-to-real reinforcement learning (RL) framework that utilizes 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 to bridge the actuation gaps with randomization of non-ideal effects such as backlash, torque-speed saturation. Using an asymmetric actor-critic PPO pipeline trained entirely in simulation, our policies deploy directly to a five-finger hand. The resulting policies demonstrated two essential skills: (1) command-based, controllable grasp force tracking, and (2) reorientation of objects in the hand, both of which were robustly executed without fine-tuning on the robot. By combining tactile and torque in the observation space with effective sensing/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.

Problem

• sim-to-real
• dexterous manipulation
• force-based grasping
• reality gap
• zero-shot transfer

Innovation

• sim-to-real
• tactile simulation
• zero-shot transfer
• dexterous manipulation
• torque estimation