ROMay 1

Learning while Deploying: Fleet-Scale Reinforcement Learning for Generalist Robot Policies

arXiv:2605.0041695.6h-index: 4
Predicted impact top 5% in RO · last 90 daysOriginality Incremental advance
AI Analysis

For robot learning, LWD provides a practical framework to improve generalist policies during deployment, addressing distribution shifts and long-tail failures.

LWD enables continual post-training of generalist robot policies using fleet-scale reinforcement learning, achieving 95% average success rate across eight real-world tasks with largest gains on long-horizon tasks.

Generalist robot policies increasingly benefit from large-scale pretraining, but offline data alone is insufficient for robust real-world deployment. Deployed robots encounter distribution shifts, long-tail failures, task variations, and human correction opportunities that fixed demonstration datasets cannot fully capture. We present Learning While Deploying (LWD), a fleet-scale offline-to-online reinforcement learning framework for continual post-training of generalist Vision-Language-Action (VLA) policies. Starting from a pretrained VLA policy, LWD closes the loop between deployment, shared physical experience, policy improvement, and redeployment by using autonomous rollouts and human interventions collected across a robot fleet. To stabilize learning from heterogeneous, sparse-reward fleet data, LWD combines Distributional Implicit Value Learning (DIVL) for robust value estimation with Q-learning via Adjoint Matching (QAM) for policy extraction in flow-based VLA action generators. We validate LWD on a fleet of 16 dual-arm robots across eight real-world manipulation tasks, including semantic grocery restocking and 3--5 minute long-horizon tasks. A single generalist policy improves as fleet experience accumulates, reaching an average success rate of 95%, with the largest gains on long-horizon tasks.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes