FL-IA3C: A Fuzzy Logic-Based Improved A3C Algorithm for Autonomous Vehicle Obstacle Avoidance and Path Planning.

Saved in:

Bibliographic Details
Title:	FL-IA3C: A Fuzzy Logic-Based Improved A3C Algorithm for Autonomous Vehicle Obstacle Avoidance and Path Planning.
Authors:	Qiu, Shaolin^1,2 bykjsl@163.com, Deng, Shuchao^2,3 dengsc@ahut.edu.cn, Pang, Honglei⁴ panghl@niit.edu.cn, Wang, Bo² 3181487830@qq.com, Ye, Hao⁵ 18913929898@163.com
Source:	IAENG International Journal of Computer Science. May2026, Vol. 53 Issue 5, p1716-1727. 12p.
Subjects:	Obstacle avoidance (Robotics), Fuzzy logic, Reinforcement learning, Autonomous vehicles, Uncertainty (Information theory), Robotic path planning
Abstract:	Autonomous vehicle navigation in indoor environments requires reliable obstacle avoidance and efficient path planning under significant uncertainty caused by dynamic obstacles, partial observability, and noisy sensory measurements. Conventional deep reinforcement learning methods often suffer from unstable training and performance degradation when confronted with ambiguous environmental states, while existing fuzzy reinforcement learning approaches lack effective integration with modern policy optimization mechanisms. To address these limitations, this paper proposes FL-IA3C, a fuzzy logic-enhanced Improved Asynchronous Advantage Actor-Critic framework that explicitly embeds uncertainty modeling into the policy learning process. By introducing a fuzzy inference mechanism to transform imprecise sensory inputs into structured uncertainty-aware representations and incorporating fuzzy-guided advantage modulation and entropy regulation within an improved A3C architecture, the proposed method enhances decision robustness, safety awareness, and training stability. Extensive simulation experiments in complex indoor environments demonstrate that FL-IA3C consistently achieves higher success rates, lower collision rates, improved path efficiency, and faster convergence compared with representative fuzzy and deep reinforcement learning baselines, while maintaining strong robustness under increasing sensor noise. These results validate that integrating fuzzy uncertainty modeling with asynchronous policy optimization provides an effective and principled solution for safe, efficient, and robust indoor autonomous navigation under perceptual uncertainty. [ABSTRACT FROM AUTHOR]
	Copyright of IAENG International Journal of Computer Science is the property of International Association of Engineers (IAENG) and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	Autonomous vehicle navigation in indoor environments requires reliable obstacle avoidance and efficient path planning under significant uncertainty caused by dynamic obstacles, partial observability, and noisy sensory measurements. Conventional deep reinforcement learning methods often suffer from unstable training and performance degradation when confronted with ambiguous environmental states, while existing fuzzy reinforcement learning approaches lack effective integration with modern policy optimization mechanisms. To address these limitations, this paper proposes FL-IA3C, a fuzzy logic-enhanced Improved Asynchronous Advantage Actor-Critic framework that explicitly embeds uncertainty modeling into the policy learning process. By introducing a fuzzy inference mechanism to transform imprecise sensory inputs into structured uncertainty-aware representations and incorporating fuzzy-guided advantage modulation and entropy regulation within an improved A3C architecture, the proposed method enhances decision robustness, safety awareness, and training stability. Extensive simulation experiments in complex indoor environments demonstrate that FL-IA3C consistently achieves higher success rates, lower collision rates, improved path efficiency, and faster convergence compared with representative fuzzy and deep reinforcement learning baselines, while maintaining strong robustness under increasing sensor noise. These results validate that integrating fuzzy uncertainty modeling with asynchronous policy optimization provides an effective and principled solution for safe, efficient, and robust indoor autonomous navigation under perceptual uncertainty. [ABSTRACT FROM AUTHOR]
ISSN:	1819656X