HandelBot: Real-World Piano Playing via Fast Adaptation of Dexterous Robot Policies

TL;DR

HandelBot is a robotic system that combines simulation and rapid adaptation techniques to achieve precise bimanual piano playing with minimal real-world data.

Contribution

It introduces a two-stage adaptation pipeline that refines simulation-trained policies for millimeter-scale precision in real-world piano playing.

Findings

Successfully performed five recognized songs on physical hardware.

Outperformed direct simulation deployment by 1.8 times.

Achieved high-precision control with only 30 minutes of real-world data.

Abstract

Mastering dexterous manipulation with multi-fingered hands has been a grand challenge in robotics for decades. Despite its potential, the difficulty of collecting high-quality data remains a primary bottleneck for high-precision tasks. While reinforcement learning and simulation-to-real-world transfer offer a promising alternative, the transferred policies often fail for tasks demanding millimeter-scale precision, such as bimanual piano playing. In this work, we introduce HandelBot, a framework that combines a simulation policy and rapid adaptation through a two-stage pipeline. Starting from a simulation-trained policy, we first apply a structured refinement stage to correct spatial alignments by adjusting lateral finger joints based on physical rollouts. Next, we use residual reinforcement learning to autonomously learn fine-grained corrective actions. Through extensive hardware experiments across five recognized songs, we demonstrate that HandelBot can successfully perform precise bimanual piano playing. Our system outperforms direct simulation deployment by a factor of 1.8x and requires only 30 minutes of physical interaction data.