Task-Level Robot Learning: Ball Throwing

Abstract

We are investigating how to program robots so that they learn tasks from practice. One method, task-level learning, provides advantages over simply perfecting models of the robot's lower level systems. Task-level learning can compensate for the structural modeling errors of the robot's lower level control systems and can speed up the learning process by reducing the degrees of freedoms of the models to be learned. We demonstrate two general learning procedures-fixed-model learning and refined-model learning-on a ball-throwing robot system. Both learning approaches refine the task command based on the performance error of the system, while they ignore the intermediate variables separating the lower level systems. We provide both experimental and theoretical evidence that task-level learning can improve a robot's performance of a task.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 01, 1987
Accession Number
ADA208019

Entities

People

  • Christopher G. Atkeson
  • David J. Reinkensmeyer
  • Eric Aboaf

Organizations

  • Massachusetts Institute of Technology

Tags

Communities of Interest

  • Autonomy

DTIC Thesaurus Topics

  • Adaptive Systems
  • Artificial Intelligence
  • Automation
  • Calibration
  • Computer Science
  • Computer Vision
  • Contracts
  • Control Systems
  • Control Systems Engineering
  • Eigenvalues
  • Engineering
  • Equations
  • Identification
  • Military Research
  • Numerical Analysis
  • Robotics
  • Task Performance And Analysis

Fields of Study

  • Computer science

Readers

  • Neural Network Machine Learning.
  • Robotics and Automation.
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • AI & ML - Neural Networks
  • Autonomy