Learning from Successful and Failed Demonstrations via Optimization

TL;DR

This paper introduces a novel Learning from Demonstration method that leverages both successful and failed demonstrations to improve skill learning and reproduction in robotic manipulation tasks.

Contribution

It proposes a new statistical skill model that encodes both demonstration types and finds optimal reproductions balancing success and failure data.

Findings

Successfully reproduces skills from failed demonstrations.

Outperforms existing LfD approaches in experiments.

Effective in multi-coordinate and real-world scenarios.

Abstract

Learning from Demonstration (LfD) is a popular approach that allows humans to teach robots new skills by showing the correct way(s) of performing the desired skill. Human-provided demonstrations, however, are not always optimal and the teacher usually addresses this issue by discarding or replacing sub-optimal (noisy or faulty) demonstrations. We propose a novel LfD representation that learns from both successful and failed demonstrations of a skill. Our approach encodes the two subsets of captured demonstrations (labeled by the teacher) into a statistical skill model, constructs a set of quadratic costs, and finds an optimal reproduction of the skill under novel problem conditions (i.e. constraints). The optimal reproduction balances convergence towards successful examples and divergence from failed examples. We evaluate our approach through several 2D and 3D experiments in real-world using a UR5e manipulator arm and also show that it can reproduce a skill from only failed demonstrations. The benefits of exploiting both failed and successful demonstrations are shown through comparison with two existing LfD approaches. We also compare our approach against an existing skill refinement method and show its capabilities in a multi-coordinate setting.