Screw Geometry Meets Bandits: Incremental Acquisition of Demonstrations to Generate Manipulation Plans

TL;DR

This paper introduces a method for robots to incrementally acquire demonstrations using screw geometry and bandit optimization, enabling confident task execution through active demonstration seeking.

Contribution

The paper presents a novel approach combining screw geometry and PAC-learning bandit strategies for systematic, incremental demonstration acquisition in robotic manipulation tasks.

Findings

Effective demonstration sufficiency measurement using screw geometry.

Successful active demonstration seeking in pouring and scooping tasks.

Robust confidence assessment of manipulation plans.

Abstract

In this paper, we study the problem of methodically obtaining a sufficient set of kinesthetic demonstrations, one at a time, such that a robot can be confident of its ability to perform a complex manipulation task in a given region of its workspace. Although Learning from Demonstrations has been an active area of research, the problems of checking whether a set of demonstrations is sufficient, and systematically seeking additional demonstrations have remained open. We present a novel approach to address these open problems using (i) a screw geometric representation to generate manipulation plans from demonstrations, which makes the sufficiency of a set of demonstrations measurable; (ii) a sampling strategy based on PAC-learning from multi-armed bandit optimization to evaluate the robot's ability to generate manipulation plans in a subregion of its task space; and (iii) a heuristic to seek additional demonstration from areas of weakness. Thus, we present an approach for the robot to incrementally and actively ask for new demonstration examples until the robot can assess with high confidence that it can perform the task successfully. We present experimental results on two example manipulation tasks, namely, pouring and scooping, to illustrate our approach. A short video on the method: https://youtu.be/R-qICICdEos