Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Learning-from-Observation System Considering Hardware-Level Reusability

Jun Takamatsu, Kazuhiro Sasabuchi, Naoki Wake, Atsushi Kanehira, Katsushi Ikeuchi

TL;DR

The paper tackles cross-hardware deployment challenges in robot manipulation by introducing hardware-level reusability within a learning-from-observation framework. It centers on robot-agnostic task representations and a hand-centric skill library, mapped to target robots via a general IK solver augmented with Labanotation constraints. Through two demonstrations, Place-on-Plate and Shelf, performed on Nextage and Fetch using the same task sequences, the approach verifies cross-robot applicability with minimal hand-specific changes. This work suggests that extensive reusability can be achieved by isolating robot hardware variations to a small set of features (the hand and IK) while preserving high-level task structure.

Abstract

Robot developers develop various types of robots for satisfying users' various demands. Users' demands are related to their backgrounds and robots suitable for users may vary. If a certain developer would offer a robot that is different from the usual to a user, the robot-specific software has to be changed. On the other hand, robot-software developers would like to reuse their developed software as much as possible to reduce their efforts. We propose the system design considering hardware-level reusability. For this purpose, we begin with the learning-from-observation framework. This framework represents a target task in robot-agnostic representation, and thus the represented task description can be shared with various robots. When executing the task, it is necessary to convert the robot-agnostic description into commands of a target robot. To increase the reusability, first, we implement the skill library, robot motion primitives, only considering a robot hand and we regarded that a robot was just a carrier to move the hand on the target trajectory. The skill library is reusable if we would like to the same robot hand. Second, we employ the generic IK solver to quickly swap a robot. We verify the hardware-level reusability by applying two task descriptions to two different robots, Nextage and Fetch.

Learning-from-Observation System Considering Hardware-Level Reusability

TL;DR

The paper tackles cross-hardware deployment challenges in robot manipulation by introducing hardware-level reusability within a learning-from-observation framework. It centers on robot-agnostic task representations and a hand-centric skill library, mapped to target robots via a general IK solver augmented with Labanotation constraints. Through two demonstrations, Place-on-Plate and Shelf, performed on Nextage and Fetch using the same task sequences, the approach verifies cross-robot applicability with minimal hand-specific changes. This work suggests that extensive reusability can be achieved by isolating robot hardware variations to a small set of features (the hand and IK) while preserving high-level task structure.

Abstract

Robot developers develop various types of robots for satisfying users' various demands. Users' demands are related to their backgrounds and robots suitable for users may vary. If a certain developer would offer a robot that is different from the usual to a user, the robot-specific software has to be changed. On the other hand, robot-software developers would like to reuse their developed software as much as possible to reduce their efforts. We propose the system design considering hardware-level reusability. For this purpose, we begin with the learning-from-observation framework. This framework represents a target task in robot-agnostic representation, and thus the represented task description can be shared with various robots. When executing the task, it is necessary to convert the robot-agnostic description into commands of a target robot. To increase the reusability, first, we implement the skill library, robot motion primitives, only considering a robot hand and we regarded that a robot was just a carrier to move the hand on the target trajectory. The skill library is reusable if we would like to the same robot hand. Second, we employ the generic IK solver to quickly swap a robot. We verify the hardware-level reusability by applying two task descriptions to two different robots, Nextage and Fetch.
Paper Structure (16 sections, 4 figures)

This paper contains 16 sections, 4 figures.

Table of Contents

  1. Introduction
  2. Related Work
  3. System Design for Hardware Reusability
  4. Definition of Tasks
  5. Tasks and Skill Library
  6. Hand Motion to Body Motion under Hardware-Level Reusability
  7. Implementation
  8. Testbed
  9. Observation of Demonstration
  10. Skill Library
  11. Skill Parameters on Execution
  12. Experiment
  13. Place-on-Plate Demo
  14. Shelf Demo
  15. Discussion
  16. ...and 1 more sections

Figures (4)

  • Figure 1: Two testbed robots: Nextage, Kawada Robotics and Fetch Mobile Manipulator, Fetch Robotics. They are equipped with Shadow Hand Lite, Shadow Robotics, as an end effector.
  • Figure 2: Overview of the demonstration
  • Figure 3: Place-on-plate demo. Top: Execution by Nextage. Bottom: Execution by Fetch. These are the reproduction of the demonstration in Figure \ref{['fig:demonstration']}.
  • Figure 4: Shelf demo. Top: Human demonstration. Middle: Execution by Nextage. Bottom: Execution by Fetch.