Health-Aware Model-Predictive Control of a Cooperative AGV-Based Production System

Area of Science:

• Industrial Engineering
• Robotics
• Operations Research

Background:

• Cooperating Automated Guided Vehicles (AGVs) are crucial in modern assembly systems.
• Accurate estimation of AGV battery status (state of charge and state of health) is essential for efficient operation but challenging to measure directly.
• Existing methods lack comprehensive analysis or direct online measurement capabilities.

Purpose of the Study:

• To develop a novel scheduling strategy for cooperating AGVs based on their remaining operational time.
• To propose a new state-of-charge estimator and a state-of-health predictor for AGV batteries.
• To enable predictive control of assembly processes with multiple constraints.

Main Methods:

• Developed a state-of-charge estimator using battery current and voltage sensor data, including a convergence analysis.
• Created a state-of-health predictor for AGV batteries.
• Proposed a control strategy for cooperative AGVs that allocates tasks based on previous task completion and battery operational time.

Main Results:

• The proposed state-of-charge estimator provides accurate online battery status.
• The state-of-health predictor enables proactive maintenance and operational planning.
• The new scheduling strategy effectively manages task allocation for cooperating AGVs under various constraints.
• Experimental studies validated the performance of the proposed approach in a simulated assembly environment.

Conclusions:

• The developed methods enable accurate online estimation of AGV battery state of charge and health.
• The novel scheduling strategy enhances the predictive control capabilities of cooperative AGV systems.
• The approach offers a robust solution for optimizing assembly processes with complex operational constraints.