LAGO Policy: Latency-Aware Asynchronous Diffusion Policies with Goal-Directed Collision-Free Planning for Smooth Manipulation

2026-06-16

AI Summary

This work addresses the challenges of asynchronous inference in diffusion-based policies, which often suffer from discontinuous actions and a lack of collision-avoidance mechanisms, leading to jitter and collisions in real-world execution. To overcome these issues, the authors propose the LAGO Policy framework, which innovatively integrates trajectory optimization with diffusion policies. The approach enhances action continuity through delay-aware classifier-free guidance, enables goal-directed collision-free planning by predicting task-interaction targets from demonstrations, and employs spatiotemporal trajectory optimization to generate smooth and dynamically feasible motion trajectories. Evaluated on a variety of complex real-world manipulation tasks, the method significantly improves task success rates while achieving high levels of motion smoothness and safety.

Abstract

Diffusion-based visuomotor policies deployed with asynchronous inference often exhibit inter-chunk discontinuities and lack explicit mechanisms for obstacle-aware execution, leading to jerky motions and collisions that hinder reliable manipulation in real-world scenes. To address these issues, we propose LAGO Policy, a unified asynchronous action-generation framework that integrates trajectory optimization with diffusion policy for smooth and safe execution. LAGO Policy improves inter-chunk consistency via latency-aware classifier-free guidance conditioning on future actions. It further enables goal-directed collision-free trajectory planning by predicting a task-relevant interaction goal from demonstrations. Finally, spatial-temporal trajectory optimization refines the actions to be executed for low-jerk and feasible motion. Extensive real-world experiments demonstrate that LAGO Policy achieves smooth collision-free execution with high task success across challenging manipulation tasks. Project Website: https://lago-policy.github.io/

Problem

• asynchronous diffusion policies
• inter-chunk discontinuities
• collision-free planning
• smooth manipulation
• latency-aware control

Innovation

• Latency-Aware Diffusion
• Asynchronous Policy
• Collision-Free Planning
• Trajectory Optimization
• Goal-Directed Manipulation