ワンACTプレイ:単一実演からの行動クローニングと行動チャンキングトランスフォーマー
One ACT Play: Single Demonstration Behavior Cloning with Action Chunking Transformers
単一の人間実演を線形変換で拡張し、行動クローニングでブロック操作タスクを学習させる手法を提案。推論時の時間アンサンブルに予測の標準偏差を取り入れ、環境変化への頑健性を向上させた。
著者: Abraham George, Amir Barati Farimani
分類: cs.RO, cs.AI, cs.LG
原文アブストラクト
Learning from human demonstrations (behavior cloning) is a cornerstone of robot learning. However, most behavior cloning algorithms require a large number of demonstrations to learn a task, especially for general tasks that have a large variety of initial conditions. Humans, however, can learn to complete tasks, even complex ones, after only seeing one or two demonstrations. Our work seeks to emulate this ability, using behavior cloning to learn a task given only a single human demonstration. We achieve this goal by using linear transforms to augment the single demonstration, generating a set of trajectories for a wide range of initial conditions. With these demonstrations, we are able to train a behavior cloning agent to successfully complete three block manipulation tasks. Additionally, we developed a novel addition to the temporal ensembling method used by action chunking agents during inference. By incorporating the standard deviation of the action predictions into the ensembling method, our approach is more robust to unforeseen changes in the environment, resulting in significant performance improvements.
