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

Efficient evaluation functions for evolving coordination.

A Agogino1, K Tumer

  • 1UCSC-NASA Ames Research Center, Moffett Field, CA 94035, USA. Adrian.K.Agogino@nasa.gov

Evolutionary Computation
|June 17, 2008
PubMed
Summary
This summary is machine-generated.

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

Ensembles of radial basis function networks for spectroscopic detection of cervical precancer.

IEEE transactions on bio-medical engineering·1998
Same journal

Computing Optimal Populations for Binary Problems using Logic Minimization.

Evolutionary computation·2026
Same journal

Enhancing Generalization and Scalability for Multi-Objective Optimization with Population Pre-Training.

Evolutionary computation·2026
Same journal

XCS for Sequential Perceptual Aliasing in Multi-Step Decision Making.

Evolutionary computation·2026
Same journal

A dynamic multi-objective evolutionary algorithm using dual-space prediction and surrogate-based sampling.

Evolutionary computation·2026
Same journal

Adapting MOEA/D to CMA-ES for Dealing with Ill-conditioned Multiobjective Problems.

Evolutionary computation·2026
Same journal

Editorial of the Special Issue: Parallel Problem Solving from Nature PPSN 2024 Extended Versions of Best Paper Candidates.

Evolutionary computation·2026
See all related articles

This study introduces new fitness functions for evolving coordinated control in large systems. These functions improve multi-component coordination, especially in challenging dynamic and noisy environments.

Area of Science:

  • Artificial Intelligence
  • Robotics
  • Control Systems Engineering

Background:

  • Evolving control policies for large multi-component systems is computationally challenging due to vast search spaces.
  • Directly evolving system-wide policies often yields unsatisfactory results in dynamic and stochastic environments.
  • Decentralized approaches require careful design of component-level evaluation functions to ensure global coordination.

Purpose of the Study:

  • To develop novel fitness evaluation functions for efficiently evolving coordination in large multi-component systems.
  • To address the challenges of alignment and sensitivity in component-level fitness functions for decentralized control.
  • To demonstrate the effectiveness of these functions in a complex multi-rover coordination task.

Main Methods:

Related Experiment Videos

  • Proposed a method for evolving individual system components with tailored fitness functions.
  • Designed component evaluation functions that are aligned with global objectives and sensitive to individual component performance.
  • Applied these functions to a distributed control problem involving multiple rovers maximizing information collection in dynamic, noisy, and communication-limited environments.

Main Results:

  • The evolved control policies using aligned and component-sensitive functions outperformed global evaluation functions by up to 400% in a multi-rover coordination task.
  • Performance improvements increased with problem difficulty (larger systems, increased noise, reduced communication).
  • Analysis quantified the alignment and sensitivity characteristics, validating the systematic study of the fitness functions.

Conclusions:

  • Aligned and component-sensitive fitness functions are crucial for effective decentralized control in large, complex systems.
  • This approach enables efficient evolution of coordinated behaviors in challenging environments where global optimization is intractable.
  • The developed methodology offers a scalable solution for coordinating multiple agents in dynamic and uncertain conditions.