Task-Adaptive Robot Learning from Demonstration with Gaussian Process Models under Replication
This work addresses the problem of task adaptation in robot learning from demonstration, which is incremental as it builds on existing GP methods by incorporating task parameters and replication structure.
The paper tackles the challenge of enabling robots to adapt learned skills from demonstrations to different scenarios by using Gaussian Process models with task parameters and exploiting replications in data, resulting in significantly reduced computational costs for model fitting in complex tasks.
Learning from Demonstration (LfD) is a paradigm that allows robots to learn complex manipulation tasks that can not be easily scripted, but can be demonstrated by a human teacher. One of the challenges of LfD is to enable robots to acquire skills that can be adapted to different scenarios. In this paper, we propose to achieve this by exploiting the variations in the demonstrations to retrieve an adaptive and robust policy, using Gaussian Process (GP) models. Adaptability is enhanced by incorporating task parameters into the model, which encode different specifications within the same task. With our formulation, these parameters can be either real, integer, or categorical. Furthermore, we propose a GP design that exploits the structure of replications, i.e., repeated demonstrations with identical conditions within data. Our method significantly reduces the computational cost of model fitting in complex tasks, where replications are essential to obtain a robust model. We illustrate our approach through several experiments on a handwritten letter demonstration dataset.