DFL-TORO: A One-Shot Demonstration Framework for Learning Time-Optimal Robotic Manufacturing Tasks

TL;DR

DFL-TORO introduces a one-shot kinesthetic demonstration framework that optimizes robotic manufacturing tasks for time efficiency and smoothness, reducing demonstration requirements and improving robot operation in industrial settings.

Contribution

The paper presents a novel one-shot demonstration framework with an optimization-based smoothing algorithm for time-optimal, jerk-regulated robotic task learning from minimal demonstrations.

Findings

Significant reduction in demonstration noise and improved efficiency.

Successful application on FR3 and ABB YuMi robots in manufacturing tasks.

Enhanced learning outcomes using Dynamic Movement Primitives (DMPs).

Abstract

This paper presents DFL-TORO, a novel Demonstration Framework for Learning Time-Optimal Robotic tasks via One-shot kinesthetic demonstration. It aims at optimizing the process of Learning from Demonstration (LfD), applied in the manufacturing sector. As the effectiveness of LfD is challenged by the quality and efficiency of human demonstrations, our approach offers a streamlined method to intuitively capture task requirements from human teachers, by reducing the need for multiple demonstrations. Furthermore, we propose an optimization-based smoothing algorithm that ensures time-optimal and jerk-regulated demonstration trajectories, while also adhering to the robot's kinematic constraints. The result is a significant reduction in noise, thereby boosting the robot's operation efficiency. Evaluations using a Franka Emika Research 3 (FR3) robot for a variety of tasks further substantiate the efficacy of our framework, highlighting its potential to transform kinesthetic demonstrations in contemporary manufacturing environments. Moreover, we take our proposed framework into a real manufacturing setting operated by an ABB YuMi robot and showcase its positive impact on LfD outcomes by performing a case study via Dynamic Movement Primitives (DMPs).