Inferring the Geometric Nullspace of Robot Skills from Human Demonstrations

TL;DR

This paper introduces a framework that learns robot skills from human demonstrations by modeling geometric nullspaces and constraints, enabling robots to execute learned skills in various environments.

Contribution

The paper presents a novel method to infer geometric nullspaces and constraints from human demonstrations for robot skill learning and execution.

Findings

Successfully learned skills from human demonstrations.

Executed skills on simulated and real robots.

Demonstrated adaptability to different environments.

Abstract

In this paper we present a framework to learn skills from human demonstrations in the form of geometric nullspaces, which can be executed using a robot. We collect data of human demonstrations, fit geometric nullspaces to them, and also infer their corresponding geometric constraint models. These geometric constraints provide a powerful mathematical model as well as an intuitive representation of the skill in terms of the involved objects. To execute the skill using a robot, we combine this geometric skill description with the robot's kinematics and other environmental constraints, from which poses can be sampled for the robot's execution. The result of our framework is a system that takes the human demonstrations as input, learns the underlying skill model, and executes the learnt skill with different robots in different dynamic environments. We evaluate our approach on a simulated industrial robot, and execute the final task on the iCub humanoid robot.