On Learning, Representing, and Generalizing a Task in a Humanoid Robot

TLDR

The paper proposes a programming‑by‑demonstration framework that extracts task features and generalizes learned skills to new contexts. The framework is validated by having a human demonstrate simple manipulation tasks to a humanoid robot, projecting motion data into a latent space with PCA, encoding it with Gaussian/Bernoulli mixture models, and using Gaussian mixture regression to compute an optimal trajectory that maximizes an imitation metric. The approach produces a spatio‑temporal correlation metric that quantifies imitation performance.

Abstract

We present a programming-by-demonstration framework for generically extracting the relevant features of a given task and for addressing the problem of generalizing the acquired knowledge to different contexts. We validate the architecture through a series of experiments, in which a human demonstrator teaches a humanoid robot simple manipulatory tasks. A probability-based estimation of the relevance is suggested by first projecting the motion data onto a generic latent space using principal component analysis. The resulting signals are encoded using a mixture of Gaussian/Bernoulli distributions (Gaussian mixture model/Bernoulli mixture model). This provides a measure of the spatio-temporal correlations across the different modalities collected from the robot, which can be used to determine a metric of the imitation performance. The trajectories are then generalized using Gaussian mixture regression. Finally, we analytically compute the trajectory which optimizes the imitation metric and use this to generalize the skill to different contexts