Learning Periodic Tasks from Human Demonstrations
This addresses the challenge of teaching robots periodic tasks with deformable objects like cloth and granular matter, which is incremental as it builds on existing methods for visual correspondence and movement primitives.
The paper tackles the problem of learning periodic tasks from visual demonstrations by leveraging periodicity in policy structure and using active learning to optimize rhythmic dynamic movement primitives, achieving optimization from a single human video demonstration within few robot trials.
We develop a method for learning periodic tasks from visual demonstrations. The core idea is to leverage periodicity in the policy structure to model periodic aspects of the tasks. We use active learning to optimize parameters of rhythmic dynamic movement primitives (rDMPs) and propose an objective to maximize the similarity between the motion of objects manipulated by the robot and the desired motion in human video demonstrations. We consider tasks with deformable objects and granular matter whose states are challenging to represent and track: wiping surfaces with a cloth, winding cables/wires, and stirring granular matter with a spoon. Our method does not require tracking markers or manual annotations. The initial training data consists of 10-minute videos of random unpaired interactions with objects by the robot and human. We use these for unsupervised learning of a keypoint model to get task-agnostic visual correspondences. Then, we use Bayesian optimization to optimize rDMPs from a single human video demonstration within few robot trials. We present simulation and hardware experiments to validate our approach.