StageCraft: Execution Aware Mitigation of Distractor and Obstruction Failures in VLA Models

TL;DR

StageCraft is a training-free, VLM-based method that improves vision-language action models by manipulating initial environments to mitigate failures caused by distractors and obstructions, significantly enhancing real-world task performance.

Contribution

We introduce StageCraft, a novel environment manipulation approach leveraging large vision-language models to improve policy robustness without additional training.

Findings

40% performance improvement in real-world tasks

Effective in diverse distractor and obstruction scenarios

Adapts intervention based on policy strength

Abstract

Large scale pre-training on text and image data along with diverse robot demonstrations has helped Vision Language Action models (VLAs) to generalize to novel tasks, objects and scenes. However, these models are still susceptible to failure in the presence of execution-time impediments such as distractors and physical obstructions in the robot's workspace. Existing policy improvement methods finetune base VLAs to improve generalization, yet they still struggle in unseen distractor settings. To address this problem, we investigate whether internet-scale pretraining of large vision-language models (VLMs) can be leveraged to reason about these impediments and mitigate policy failures. To this end, we propose StageCraft, a training-free approach to improve pretrained VLA policy performance by manipulating the environment's initial state using VLM-based in-context reasoning. StageCraft takes policy rollout videos and success labels as input and leverages VLM's reasoning ability to infer which objects in the initial state need to be manipulated to avoid anticipated execution failures. StageCraft is an extensible plug-and-play module that does not introduce additional constraints on the underlying policy, and only requires a few policy rollouts to work. We evaluate performance of state-of-the-art VLA models with StageCraft and show an absolute 40% performance improvement across three real world task domains involving diverse distractors and obstructions. Our simulation experiments in RLBench empirically show that StageCraft tailors its extent of intervention based on the strength of the underlying policy and improves its performance with more in-context samples. Videos of StageCraft in effect can be found at https://stagecraft-decorator.github.io/stagecraft/ .