Action ControlNet: A Lightweight Delay-Aware Adapter for Smooth Asynchronous Control in Vision-Language-Action Models
For robot manipulation using VLA models, ACNet addresses the problem of action discontinuities and jitter due to inference delay without requiring full-policy retraining or architecture-specific runtime logic.
Action ControlNet (ACNet) is a lightweight delay-aware adapter that improves smoothness and robustness of asynchronous control in VLA models by using executed motion suffix as a residual condition, achieving smoother trajectories than direct chunk stitching on Kinetix, Meta-World MT50, and a real-world SO-ARM101 platform.
Vision-language-action (VLA) models have shown strong potential for general-purpose robot manipulation, but their inference latency remains a major obstacle to stable high-frequency control. Asynchronous execution mitigates this bottleneck by overlapping policy inference with action execution, yet the next action chunk is still predicted from stale observations while the robot continues to move. Direct chunk stitching therefore introduces handoff discontinuities, action jitter, and failures in contact-rich manipulation. Existing remedies typically require either full-policy retraining or architecture-specific runtime logic. This work proposes Action ControlNet (ACNet), a lightweight delay-aware adapter that uses the executed motion suffix as a residual condition for a mostly frozen action head. ACNet leaves the pretrained backbone unchanged, introduces few trainable parameters, and remains compatible with generative action heads such as diffusion and flow matching. On Kinetix, Meta-World MT50, and a real-world SO-ARM101 platform, ACNet improves robustness under inference delay and yields smoother asynchronous trajectories than direct chunk stitching, while remaining more lightweight than full delay-conditioned retraining.