Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Autonomous learning based on cost assumptions: theoretical studies and experiments in robot control.

C H Ribeiro1, E M Hemerly

  • 1Technological Institute of Aeronautics, São José dos Campos-SP, Brazil.

International Journal of Neural Systems
|November 24, 1999
PubMed
Summary

This study introduces a generalized Q-learning method for autonomous learning systems. It enables efficient information sharing to improve learning speed and safely incorporates prior knowledge for robot navigation tasks.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Adaptive observers for unknown general nonlinear systems.

IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2008
Same author

Adaptive control for mobile robot using wavelet networks.

IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2008
Same author

[Meningiomas of the cavernous sinus: the surgical resectability and complications].

Arquivos de neuro-psiquiatria·2001
Same author

Autonomous learning based on cost assumptions: theoretical studies and experiments in robot control.

International journal of neural systems·2000
Same author

[Subfrontal approach in sellar and suprasellar lesions].

Arquivos de neuro-psiquiatria·2000

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Robotics

Background:

  • Autonomous learning relies on experience, which is often time-consuming in real-world applications.
  • Classical learning algorithms struggle with the crude information available from system dynamics.
  • There's a need to embed facilitating information without altering core algorithm structures.

Purpose of the Study:

  • To develop a generalized Q-learning formulation for enhanced autonomous learning.
  • To enable information spreading from single updates to neighboring states.
  • To provide a mechanism for safely embedding prior knowledge into learning algorithms.

Main Methods:

  • Investigated a generalized formulation of the Q-learning method.
  • Developed a mechanism for spreading information to neighboring states.

Related Experiment Videos

  • Demonstrated the approach in a modified reinforcement learning algorithm for robot navigation.

    • Main Results:

    • The proposed method converges to optimality.
    • It allows for the safe embedding of prior knowledge about the state space structure.
    • Successfully applied to a real robot navigation task.

    Conclusions:

    • The generalized Q-learning formulation enhances autonomous learning efficiency.
    • This approach facilitates the integration of prior knowledge, improving performance in complex tasks.
    • The method is effective for real-world applications like robot navigation.