CoLA-Flow Policy: Temporally Coherent Imitation Learning via Continuous Latent Action Flow Matching for Robotic Manipulation

TL;DR

The paper introduces CoLA-Flow Policy, a novel imitation learning framework that encodes action sequences into latent trajectories, enabling fast, stable, and long-horizon robotic manipulation with improved robustness and efficiency.

Contribution

It proposes a trajectory-level flow matching approach in a continuous latent space, decoupling motion structure from control noise for enhanced long-horizon robotic control.

Findings

Achieves near-single-step inference for robotic manipulation.

Improves trajectory smoothness by up to 93.7%.

Increases task success rate by up to 25 percentage points.

Abstract

Learning long-horizon robotic manipulation requires jointly achieving expressive behavior modeling, real-time inference, and stable execution, which remains challenging for existing generative policies. Diffusion-based approaches offer strong modeling capacity but incur high inference latency, while flow matching enables fast, near-single-step generation yet often suffers from unstable execution when operating directly in the raw action space. We propose Continuous Latent Action Flow Policy (CoLA-Flow Policy), a trajectory-level imitation learning framework that performs flow matching in a continuous latent action space. By encoding action sequences into temporally coherent latent trajectories and learning an explicit latent-space flow, CoLA-Flow Policy decouples global motion structure from low-level control noise, enabling smooth and reliable long-horizon execution. The framework further integrates geometry-aware point cloud conditioning and execution-time multimodal modulation, using visual cues as a representative modality to enhance real-world robustness. Experiments in simulation and on real robots show that CoLA-Flow Policy achieves near-single-step inference, improves trajectory smoothness by up to 93.7% and task success by up to 25 percentage points over raw action-space flow baselines, while remaining significantly faster than diffusion-based policies.