A Generic Multi-Layer Architecture Based on ROS-JADE Integration for Autonomous Transport Vehicles

Area of Science:

• Manufacturing Systems Engineering
• Robotics
• Artificial Intelligence

Background:

• Manufacturing systems require reconfiguration due to dynamic changes like equipment failure or urgent orders.
• Autonomous Transport Vehicles (ATVs) are crucial for flexible and decentralized manufacturing.
• Existing robotic frameworks for ATVs lack the social abilities needed for intelligent interaction with other systems.

Purpose of the Study:

• To develop a generic multi-layer architecture enabling social abilities for ATVs.
• To facilitate intelligent interaction among heterogeneous Cyber Physical Production Systems for manufacturing reconfiguration.
• To integrate robotic frameworks (RFs) with Multi-Agent Systems (MAS) for enhanced ATV functionality.

Main Methods:

• Proposed a novel multi-layer architecture integrating a robotic framework (RF) with a Multi-Agent System (MAS).
• Implemented the architecture using ROS (Robotic Operating System) and JADE (Java Agent DEvelopment Framework).
• Focused on enabling peer-to-peer negotiation and decision-making capabilities for ATVs.

Main Results:

• Successfully integrated a RF with a MAS to provide social abilities to ATVs.
• Demonstrated the architecture's capability for intelligent interaction in flexible manufacturing.
• The implementation on ROS and JADE validates the proposed approach.

Conclusions:

• The developed architecture provides a holistic solution for intelligent interaction of transportation systems in flexible manufacturing.
• This integration enhances ATVs' ability to participate in dynamic manufacturing reconfiguration.
• This work represents a significant advancement in creating adaptable and intelligent factory environments.