Learning from demonstration with model-based Gaussian process
This work addresses robot learning from demonstrations for improved adaptability in tasks with variable human inputs, representing an incremental advancement in the field.
The paper tackles the problem of adapting robot behavior based on variability in human demonstrations and task uncertainty, proposing a multi-output Gaussian process method that enables precise tracking of via-points while maintaining compliance in high-variability regions, validated in simulated and real-robot experiments.
In learning from demonstrations, it is often desirable to adapt the behavior of the robot as a function of the variability retrieved from human demonstrations and the (un)certainty encoded in different parts of the task. In this paper, we propose a novel multi-output Gaussian process (MOGP) based on Gaussian mixture regression (GMR). The proposed approach encapsulates the variability retrieved from the demonstrations in the covariance of the MOGP. Leveraging the generative nature of GP models, our approach can efficiently modulate trajectories towards new start-, via- or end-points defined by the task. Our framework allows the robot to precisely track via-points while being compliant in regions of high variability. We illustrate the proposed approach in simulated examples and validate it in a real-robot experiment.