PILOT: Guiding Long-Horizon World-Model Planning for Robot Manipulation with Visual Subgoals and Policy Priors
PILOT: Guiding Long-Horizon World-Model Planning for Robot Manipulation with Visual Subgoals and Policy Priors
The paper presents PILOT, a method that enhances long-horning robotic manipulation by integrating visual subgoals and policy priors. Using a video model to generate intermediate goals and a learned policy to initialize the cross-entropy method (CEM), PILOT achieves 68.8% success in real-robot tasks, outperforming baselines like direct final-goal planning and policy-only execution. Simulation results show consistent gains over longer horizons, with improvements in both task execution and planning efficiency.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- paper: not_reported
- code: not_reported
- weights: not_reported
- dataset: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
68.8% success, compared with 55.0% without policy initialization, 13.8% for policy-only execution, and 0.0% for direct final-goal planning. Simulation results show that these gains persist over longer horizons: PILOT exceeds direct execution of the same policy by 30 percentage points on long Push-T tasks and by 22 percentage points on LIBERO-10. Policy initialization also accelerates local optimization: on Push-T, C
Open source S4
Source：arXiv Robotics — research abstracts · arxiv.org