Tac2Real: Reliable and GPU Visuotactile Simulation for Online Reinforcement Learning and Zero-Shot Real-World Deployment

TL;DR

Tac2Real is a high-performance visuotactile simulation framework that enables efficient online reinforcement learning and reliable zero-shot transfer to real-world robotic contact tasks.

Contribution

We introduce Tac2Real, a lightweight, GPU-accelerated visuotactile simulation framework with domain gap reduction techniques for zero-shot sim-to-real transfer.

Findings

High success rate in real-world peg insertion after zero-shot transfer

Efficient online RL training enabled by high-throughput parallel simulation

Effective domain gap narrowing with TacAlign

Abstract

Visuotactile sensors are indispensable for contact-rich robotic manipulation tasks. However, policy learning with tactile feedback in simulation, especially for online reinforcement learning (RL), remains a critical challenge, as it demands a delicate balance between physics fidelity and computational efficiency. To address this challenge, we present Tac2Real, a lightweight visuotactile simulation framework designed to enable efficient online RL training. Tac2Real integrates the Preconditioned Nonlinear Conjugate Gradient Incremental Potential Contact (PNCG-IPC) method with a multi-node, multi-GPU high-throughput parallel simulation architecture, which can generate marker displacement fields at interactive rates. Meanwhile, we propose a systematic approach, TacAlign, to narrow both structured and stochastic sources of domain gap, ensuring a reliable zero-shot sim-to-real transfer. We further evaluate Tac2Real on the contact-rich peg insertion task. The zero-shot transfer results achieve a high success rate in the real-world scenario, verifying the effectiveness and robustness of our framework. The project page is: https://ningyurichard.github.io/tac2real-project-page/