Towards Robot Task Planning From Probabilistic Models of Human Skills

TL;DR

This paper presents a probabilistic approach to robot motion planning that learns from expert demonstrations to enable robots to perform complex manipulation tasks in new environments.

Contribution

It introduces a method to represent robot skills as probability distributions over features learned from demonstrations, facilitating robust generalization and task execution.

Findings

Successfully applied to diverse case studies including games, assembly, and real robot experiments.

Demonstrated effective generalization of skills to new environments.

Enabled complex task execution through probabilistic motion planning.

Abstract

We describe an algorithm for motion planning based on expert demonstrations of a skill. In order to teach robots to perform complex object manipulation tasks that can generalize robustly to new environments, we must (1) learn a representation of the effects of a task and (2) find an optimal trajectory that will reproduce these effects in a new environment. We represent robot skills in terms of a probability distribution over features learned from multiple expert demonstrations. When utilizing a skill in a new environment, we compute feature expectations over trajectory samples in order to stochastically optimize the likelihood of a trajectory in the new environment. The purpose of this method is to enable execution of complex tasks based on a library of probabilistic skill models. Motions can be combined to accomplish complex tasks in hybrid domains. Our approach is validated in a variety of case studies, including an Android game, simulated assembly task, and real robot experiment with a UR5.