A Hybrid Approach Using Vector Field Histogram and Deep Reinforcement Learning for Dynamic Path Planning.

Saved in:

Bibliographic Details
Title:	A Hybrid Approach Using Vector Field Histogram and Deep Reinforcement Learning for Dynamic Path Planning.
Authors:	Hameed Qasim, Maysam¹ pgs.maysam.qasm@uobasrah.edu.iq, Al-Darraji, Salah¹
Source:	Iraqi Journal for Electrical & Electronic Engineering. Jun2026, Vol. 22 Issue 1, p388-400. 13p.
Subjects:	Robotic path planning, Obstacle avoidance (Robotics), Robot control systems, Reinforcement learning, Mobile robots
Abstract (English):	Autonomous mobile robots (AMRs) are becoming increasingly important in different domains such as healthcare, warehouse automation and household duties, but still encounter problems when it comes to moving around unfamiliar and dynamic environments. This study proposes an advanced robotic navigation system which combines the Soft Actor-Critic (SAC) approach and Vector Field Histogram (VFH) for path planning and avoidance obstacles in completely unknown environments. This system leverages the strengths of deep reinforcement learning and real-time obstacle detection to achieve robust and efficient navigation in certain scenarios. The SAC strategy optimizes robot navigation using policy networks and Q-networks, while the VFH method addresses obstacle avoidance by sensor data processing and dynamically adjusting the robot’s angular velocity to avoid collision. For testing and implementing this system, Gazebo simulation and Robot Operating System (ROS) are used. Experimental results demonstrated that the proposed method outperformed the standard technique and achieved a high success rate in path planning and obstacles avoidance. [ABSTRACT FROM AUTHOR]
Abstract (Arabic):	يركز المقال على نهج هجين جديد لتخطيط المسار الديناميكي في الروبوتات المتنقلة الذاتية من خلال دمج تقنية التعلم المعزز العميق "الممثل الناعم-الناقد" (Soft Actor-Critic، SAC) مع الطريقة الكلاسيكية "مخطط الحقل الاتجاهي" (Vector Field Histogram، VFH) لتجنب العقبات. يستفيد هذا الدمج من قدرات تحسين السياسات في SAC ومن الكشف اللحظي للعقبات المعتمد على المستشعرات في VFH لتعزيز التنقل في بيئات غير معروفة وديناميكية. أظهرت التجارب المحاكاة باستخدام منصة "تيرتل بوت 3" (Turtlebot3) في بيئة "جازيبو" (Gazebo) ونظام تشغيل الروبوتات (ROS) أن الطريقة المقترحة تتفوق بشكل ملحوظ على خوارزمية SAC القياسية، محققة معدلات نجاح أعلى وتحسناً في تجنب الاصطدامات في السيناريوهات الثابتة والديناميكية على حد سواء. تسلط الدراسة الضوء على إمكانات النهج الهجين في تحسين قدرة الروبوتات المتنقلة على التكيف والسلامة والكفاءة في مهام التنقل المعقدة دون الاعتماد على خرائط بيئية مسبقة. [Extracted from the article]
	Copyright of Iraqi Journal for Electrical & Electronic Engineering is the property of Republic of Iraq Ministry of Higher Education & Scientific Research (MOHESR) 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 mobile robots (AMRs) are becoming increasingly important in different domains such as healthcare, warehouse automation and household duties, but still encounter problems when it comes to moving around unfamiliar and dynamic environments. This study proposes an advanced robotic navigation system which combines the Soft Actor-Critic (SAC) approach and Vector Field Histogram (VFH) for path planning and avoidance obstacles in completely unknown environments. This system leverages the strengths of deep reinforcement learning and real-time obstacle detection to achieve robust and efficient navigation in certain scenarios. The SAC strategy optimizes robot navigation using policy networks and Q-networks, while the VFH method addresses obstacle avoidance by sensor data processing and dynamically adjusting the robot’s angular velocity to avoid collision. For testing and implementing this system, Gazebo simulation and Robot Operating System (ROS) are used. Experimental results demonstrated that the proposed method outperformed the standard technique and achieved a high success rate in path planning and obstacles avoidance. [ABSTRACT FROM AUTHOR]
ISSN:	18145892
DOI:	10.37917/ijeee.22.1.34