Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.23087 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914573550878720 |
|---|---|
| author | Songwei, Wu Zhiduo, Jiang Wandong, Sun Guanghu, Xie Rui, Zhao Hong, Liu Yang, Liu |
| author_facet | Songwei, Wu Zhiduo, Jiang Wandong, Sun Guanghu, Xie Rui, Zhao Hong, Liu Yang, Liu |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_23087 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | CoLA-Flow Policy: Temporally Coherent Imitation Learning via Continuous Latent Action Flow Matching for Robotic Manipulation Songwei, Wu Zhiduo, Jiang Wandong, Sun Guanghu, Xie Rui, Zhao Hong, Liu Yang, Liu Robotics 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. |
| title | CoLA-Flow Policy: Temporally Coherent Imitation Learning via Continuous Latent Action Flow Matching for Robotic Manipulation |
| topic | Robotics |
| url | https://arxiv.org/abs/2601.23087 |