Saved in:
Bibliographic Details
Main Authors: Şahin, Hürkan, Dang, Van Huyen, Sayar, Erdi, Yegenoglu, Alper, Kayacan, Erdal
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.15772
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910138982465536
author Şahin, Hürkan
Dang, Van Huyen
Sayar, Erdi
Yegenoglu, Alper
Kayacan, Erdal
author_facet Şahin, Hürkan
Dang, Van Huyen
Sayar, Erdi
Yegenoglu, Alper
Kayacan, Erdal
contents Reinforcement learning (RL) often struggles in real-world tasks with high-dimensional state spaces and long horizons, where sparse or fixed rewards severely slow down exploration and cause agents to get trapped in local optima. This paper presents a fuzzy logic based reward shaping method that integrates human intuition into RL reward design. By encoding expert knowledge into adaptive and interpreable terms, fuzzy rules promote stable learning and reduce sensitivity to hyperparameters. The proposed method leverages these properties to adapt reward contributions based on the agent state, enabling smoother transitions between fast motion and precise control in challenging navigation tasks. Extensive simulation results on autonomous drone racing benchmarks show stable learning behavior and consistent task performance across scenarios of increasing difficulty. The proposed method achieves faster convergence and reduced performance variability across training seeds in more challenging environments, with success rates improving by up to approximately 5 percent compared to non fuzzy reward formulations.
format Preprint
id arxiv_https___arxiv_org_abs_2604_15772
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Fuzzy Logic Theory-based Adaptive Reward Shaping for Robust Reinforcement Learning (FARS)
Şahin, Hürkan
Dang, Van Huyen
Sayar, Erdi
Yegenoglu, Alper
Kayacan, Erdal
Robotics
Reinforcement learning (RL) often struggles in real-world tasks with high-dimensional state spaces and long horizons, where sparse or fixed rewards severely slow down exploration and cause agents to get trapped in local optima. This paper presents a fuzzy logic based reward shaping method that integrates human intuition into RL reward design. By encoding expert knowledge into adaptive and interpreable terms, fuzzy rules promote stable learning and reduce sensitivity to hyperparameters. The proposed method leverages these properties to adapt reward contributions based on the agent state, enabling smoother transitions between fast motion and precise control in challenging navigation tasks. Extensive simulation results on autonomous drone racing benchmarks show stable learning behavior and consistent task performance across scenarios of increasing difficulty. The proposed method achieves faster convergence and reduced performance variability across training seeds in more challenging environments, with success rates improving by up to approximately 5 percent compared to non fuzzy reward formulations.
title Fuzzy Logic Theory-based Adaptive Reward Shaping for Robust Reinforcement Learning (FARS)
topic Robotics
url https://arxiv.org/abs/2604.15772