Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Passivity-Based Variable Impedance Controller for Incremental Learning of Periodic Interactive Tasks

Matteo Dalle Vedove, Edoardo Lamon, Daniele Fontanelli, Luigi Palopoli, Matteo Saveriano

TL;DR

The paper presents a safety-focused, incremental learning framework for periodic interactive tasks in human–robot collaboration. It integrates a passivity-guaranteed variable impedance controller with energy tanks, online period estimation via an Adaptive Frequency Oscillator, and orientation-capable periodic Dynamic Movement Primitives to learn and execute task trajectories on the fly. An autonomy level regulator modulates learning versus execution, using energy-flow constraints to distinguish demonstrations from disturbances. Experimental validation on a wiping task with a UR5e demonstrates safe teaching-execution transitions, robust adaptation to surface changes, and controllable force during contact, highlighting practical impact for flexible, rapid reconfiguration in manufacturing.

Abstract

In intelligent manufacturing, robots are asked to dynamically adapt their behaviours without reducing productivity. Human teaching, where an operator physically interacts with the robot to demonstrate a new task, is a promising strategy to quickly and intuitively reconfigure the production line. However, physical guidance during task execution poses challenges in terms of both operator safety and system usability. In this paper, we solve this issue by designing a variable impedance control strategy that regulates the interaction with the environment and the physical demonstrations, explicitly preventing at the same time passivity violations. We derive constraints to limit not only the exchanged energy with the environment but also the exchanged power, resulting in smoother interactions. By monitoring the energy flow between the robot and the environment, we are able to distinguish between disturbances (to be rejected) and physical guidance (to be accomplished), enabling smooth and controlled transitions from teaching to execution and vice versa. The effectiveness of the proposed approach is validated in wiping tasks with a real robotic manipulator.

A Passivity-Based Variable Impedance Controller for Incremental Learning of Periodic Interactive Tasks

TL;DR

The paper presents a safety-focused, incremental learning framework for periodic interactive tasks in human–robot collaboration. It integrates a passivity-guaranteed variable impedance controller with energy tanks, online period estimation via an Adaptive Frequency Oscillator, and orientation-capable periodic Dynamic Movement Primitives to learn and execute task trajectories on the fly. An autonomy level regulator modulates learning versus execution, using energy-flow constraints to distinguish demonstrations from disturbances. Experimental validation on a wiping task with a UR5e demonstrates safe teaching-execution transitions, robust adaptation to surface changes, and controllable force during contact, highlighting practical impact for flexible, rapid reconfiguration in manufacturing.

Abstract

In intelligent manufacturing, robots are asked to dynamically adapt their behaviours without reducing productivity. Human teaching, where an operator physically interacts with the robot to demonstrate a new task, is a promising strategy to quickly and intuitively reconfigure the production line. However, physical guidance during task execution poses challenges in terms of both operator safety and system usability. In this paper, we solve this issue by designing a variable impedance control strategy that regulates the interaction with the environment and the physical demonstrations, explicitly preventing at the same time passivity violations. We derive constraints to limit not only the exchanged energy with the environment but also the exchanged power, resulting in smoother interactions. By monitoring the energy flow between the robot and the environment, we are able to distinguish between disturbances (to be rejected) and physical guidance (to be accomplished), enabling smooth and controlled transitions from teaching to execution and vice versa. The effectiveness of the proposed approach is validated in wiping tasks with a real robotic manipulator.
Paper Structure (10 sections, 20 equations, 3 figures)

This paper contains 10 sections, 20 equations, 3 figures.

Table of Contents

  1. Introduction
  2. Methodology
  3. Variable Impedance Control
  4. Incremental Learning of Periodic Tasks
  5. Online Period Estimation
  6. Autonomy Level Regulator
  7. Experiments
  8. Setup
  9. Results and Discussion
  10. Conclusion

Figures (3)

  • Figure 1:
  • Figure 2:
  • Figure 3: Snapshots taken from the demonstration (available at https://youtu.be/2fCSnqNbqz0) and collected data: (a) the human demonstrates how to clean the edge of a table, and (b) the robot executes it. Then (c), the human demonstrates a circular motion on a wooden table, and (d) the robot executes it (e), even when the object is perturbed. Finally, (f) the human shows a motion that requires the tilting of the robot end-effector, as for the poses shown in (g) and (h). Green areas represent time windows where the robot is autonomous, while orange ones represent time windows when the human is interacting.