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

Reinforcement learning algorithms for robotic navigation in dynamic environments.

Gary G Yen1, Travis W Hickey

  • 1Intelligent Systems and Control Laboratory, School of Electrical and Computer Engineering, Oklahoma State University, Stillwater, OK 74078, USA.

ISA Transactions
|April 22, 2004
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

Caformer: Rethinking Time-Series Forecasting From Causal Perspective.

IEEE transactions on cybernetics·2025
Same author

Imitation Learning for Multiobjective Optimization-AlphaMOEA.

IEEE transactions on cybernetics·2025
Same author

REaMA: Building Biomedical Relation Extraction Specialized Large Language Models Through Instruction Tuning.

IEEE transactions on neural networks and learning systems·2025
Same author

Uncovering Large Language Model Weaknesses in Character and Word Understanding and Manipulating.

IEEE transactions on neural networks and learning systems·2025
Same author

HMAMP: Designing Highly Potent Antimicrobial Peptides Using a Hypervolume-Driven Multiobjective Deep Generative Model.

Journal of medicinal chemistry·2025
Same author

Boundary-Based Active Domain Adaptation for Semantic Segmentation Under Adverse Conditions.

IEEE transactions on neural networks and learning systems·2025
Same journal

Optimization of mode discerning control for nonlinear hybrid systems subject to unknown inputs with applications to active fault diagnosis.

ISA transactions·2026
Same journal

Convergence evaluation of optimization-based stochastic iterative learning control.

ISA transactions·2026
Same journal

Adaptive utility-aware event-triggered reinforcement learning for hybrid attack scheduling against remote state estimation.

ISA transactions·2026
Same journal

Hybrid vehicle state estimation using closed-form liquid neural networks and nonlinear Kalman filtering.

ISA transactions·2026
Same journal

Cross-coupled synchronization control strategy for rebar binding robots based on impedance control.

ISA transactions·2026
Same journal

Gas flow tracking for electronic pressure control system in gas chromatography under state constraints and hysteresis：An innovative fuzzy adaptive control approach.

ISA transactions·2026
See all related articles

Reinforcement learning (RL) agents navigate dynamic environments faster with a forgetting mechanism and hierarchical structure. These improvements enhance robotic navigation performance compared to traditional methods.

Area of Science:

  • Robotics
  • Artificial Intelligence
  • Machine Learning

Background:

  • Reinforcement learning (RL) algorithms are crucial for autonomous systems.
  • Dynamic environments pose significant challenges for traditional RL agents in robotic navigation.
  • Enhancing RL agent adaptability and efficiency is key for real-world applications.

Purpose of the Study:

  • To investigate improvements for reinforcement learning (RL) algorithms in dynamic environments.
  • To evaluate RL algorithms applied to robotic navigation tasks.
  • To enhance the performance and learning speed of RL agents.

Main Methods:

  • Simulations were conducted to assess proposed RL algorithm enhancements.
  • Evaluated improvements include: a forgetting mechanism, feature-based state inputs, and hierarchical structuring.

Related Experiment Videos

  • Compared proposed methods against established techniques and optimal solutions.

    • Main Results:

    • A forgetting mechanism significantly reduced learning times for RL agents in dynamic settings.
    • Feature-based state inputs alone did not improve performance due to a lack of positional awareness.
    • Hierarchical RL agent structures, combining feature-based and standard agents, significantly boosted performance.

    Conclusions:

    • A forgetting mechanism and hierarchical structuring substantially improve RL agent performance in dynamic environments.
    • Hierarchical RL agents integrating feature-based obstacle avoidance and standard navigation show superior results.
    • These enhancements offer a significant advantage over traditional RL approaches for robotic navigation.