Predicting and executing a sequence of actions without intermediate replanning, which is known as action chunking, is increasingly used in robot learning from human demonstrations.
This approach involves two key strategies:
-
it predicts multi-step action sequences and executes them either fully or partially
-
it models the distribution of action chunks and performs sampling from the learned model, either independently or with weak dependencies.
Action chunking allows the learner to better capture temporal dependencies in demonstrations and generate more consistent and stable actions.