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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Active Inference with Dynamic Planning and Information Gain in Continuous Space by Inferring Low-Dimensional Latent States.

Entropy (Basel, Switzerland)·2025
Same author

The Epistemic Uncertainty Gradient in Spaces of Random Projections.

Entropy (Basel, Switzerland)·2025
Same author

Development of compositionality through interactive learning of language and action of robots.

Science robotics·2025
Same author

Synergizing habits and goals with variational Bayes.

Nature communications·2024
Same author

Incremental Learning of Goal-Directed Actions in a Dynamic Environment by a Robot Using Active Inference.

Entropy (Basel, Switzerland)·2023
Same author

Turn-Taking Mechanisms in Imitative Interaction: Robotic Social Interaction Based on the Free Energy Principle.

Entropy (Basel, Switzerland)·2023

Related Experiment Video

Updated: Sep 26, 2025

SSVEP-based Experimental Procedure for Brain-Robot Interaction with Humanoid Robots
11:01

SSVEP-based Experimental Procedure for Brain-Robot Interaction with Humanoid Robots

Published on: November 24, 2015

13.3K

Goal-Directed Planning and Goal Understanding by Extended Active Inference: Evaluation through Simulated and Physical

Takazumi Matsumoto1, Wataru Ohata1, Fabien C Y Benureau1

  • 1Cognitive Neurorobotics Research Unit, Okinawa Institute of Science and Technology, Okinawa 904-0495, Japan.

Entropy (Basel, Switzerland)
|April 23, 2022
PubMed
Summary

This study introduces a new model for goal-directed action planning and generation by extending active inference. The model enables agents to understand and generate plans for various goals, improving robotic task performance.

Keywords:
active inferencegoal-directed action planningteleology

More Related Videos

The Modular Design and Production of an Intelligent Robot Based on a Closed-Loop Control Strategy
11:53

The Modular Design and Production of an Intelligent Robot Based on a Closed-Loop Control Strategy

Published on: October 14, 2017

11.8K
Robotic Sensing and Stimuli Provision for Guided Plant Growth
08:02

Robotic Sensing and Stimuli Provision for Guided Plant Growth

Published on: July 1, 2019

8.1K

Related Experiment Videos

Last Updated: Sep 26, 2025

SSVEP-based Experimental Procedure for Brain-Robot Interaction with Humanoid Robots
11:01

SSVEP-based Experimental Procedure for Brain-Robot Interaction with Humanoid Robots

Published on: November 24, 2015

13.3K
The Modular Design and Production of an Intelligent Robot Based on a Closed-Loop Control Strategy
11:53

The Modular Design and Production of an Intelligent Robot Based on a Closed-Loop Control Strategy

Published on: October 14, 2017

11.8K
Robotic Sensing and Stimuli Provision for Guided Plant Growth
08:02

Robotic Sensing and Stimuli Provision for Guided Plant Growth

Published on: July 1, 2019

8.1K

Area of Science:

  • Artificial Intelligence
  • Robotics
  • Computational Neuroscience

Background:

  • Goal-directed behavior is crucial for intelligent agents.
  • Existing frameworks struggle to integrate planning and understanding of diverse goals.
  • Active inference provides a promising but incomplete framework for goal-directed action.

Purpose of the Study:

  • To extend the active inference framework for goal-directed action planning and generation.
  • To develop a model capable of specifying and understanding both static and dynamic goals.
  • To enable agents to infer current states and generate future action plans based on observations.

Main Methods:

  • Utilized a variational recurrent neural network (VRNN) model.
  • Extended the active inference framework to incorporate teleological principles.
  • Integrated goal specification for static sensory states and dynamic processes.
  • Developed state estimation from past sensory observations for future planning.

Main Results:

  • Demonstrated goal-directed action planning and generation capabilities.
  • Showcased the model's ability to understand goals from sensory input.
  • Successfully inferred current states to generate future action plans.
  • Validated the model on a simulated mobile agent and a real humanoid robot.

Conclusions:

  • The extended active inference model effectively formulates goal-directed action planning and generation.
  • The model's flexibility in goal specification enhances its applicability in robotics and AI.
  • The approach offers a unified framework for action planning, goal understanding, and state inference.