FreqPolicy: Frequency Autoregressive Visuomotor Policy with Continuous Tokens

TL;DR

FreqPolicy introduces a hierarchical frequency-domain autoregressive approach with continuous tokens for visuomotor policy learning, improving robotic manipulation accuracy and efficiency by capturing motion structure effectively.

Contribution

It proposes a novel frequency-based hierarchical modeling paradigm with continuous latent representations for visuomotor policies, advancing beyond existing methods.

Findings

Outperforms existing methods in accuracy and efficiency

Effective modeling of motion structure via frequency components

Demonstrates generalization across diverse robotic tasks

Abstract

Learning effective visuomotor policies for robotic manipulation is challenging, as it requires generating precise actions while maintaining computational efficiency. Existing methods remain unsatisfactory due to inherent limitations in the essential action representation and the basic network architectures. We observe that representing actions in the frequency domain captures the structured nature of motion more effectively: low-frequency components reflect global movement patterns, while high-frequency components encode fine local details. Additionally, robotic manipulation tasks of varying complexity demand different levels of modeling precision across these frequency bands. Motivated by this, we propose a novel paradigm for visuomotor policy learning that progressively models hierarchical frequency components. To further enhance precision, we introduce continuous latent representations that maintain smoothness and continuity in the action space. Extensive experiments across diverse 2D and 3D robotic manipulation benchmarks demonstrate that our approach outperforms existing methods in both accuracy and efficiency, showcasing the potential of a frequency-domain autoregressive framework with continuous tokens for generalized robotic manipulation.Code is available at https://github.com/4DVLab/Freqpolicy