To Örebro University

Automated inspection in lithium-ion battery manufacturing has emerged as a significant enabler of quality assurance, process optimization, and operational efficiency. Defect detection at several stages from electrode manufacturing to cell packaging has a direct impact on battery safety, performance, and durability. Traditional machine vision approaches were early solutions for detecting electrode flaws, but recent improvements in deep learning, hybrid image processing, and AI assisted inspection have considerably improved accuracy, resilience, and real-time capabilities. This study proposes a comprehensive automated inspection framework that incorporates 3D structured-light profiling, high-speed 2D line-scan imaging, and edge-computing-enabled analytics throughout the production process. The system controls coating uniformity, calendering integrity, geometric precision, electrode alignment, and weld quality, all while ensuring comprehensive digital traceability via MES integration. Vision-guided robotic handling improves assembly accuracy, throughput, and process reliability. By merging multidimensional sensing modalities with AI-driven analysis, the proposed framework assures high-throughput, defect-free battery manufacture while lowering waste, boosting sustainability, and promoting Industry 5.0 digitalization.