Learning Based Artificially Assisted Manufacturing Workflow with Augmented Reality – a Comprehensive Study

The integration of Artificial Intelligence (AI) and Augmented Reality (AR) is revolutionizing modern manufacturing by enhancing automation, efficiency, and human-machine collaboration. This proposal presents a Learning-Based Artificially Assisted Manufacturing Workflow leveraging Deep Reinforcement Learning (DRL) and Transformer-Based models to optimize real-time decision-making, predictive maintenance, and process automation. The system utilizes a combination of Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) for accurate object detection, defect identification, and quality assurance within an AR-enhanced workspace. A Deep Q-Network (DQN) with Transfer Learning is proposed to adaptively optimize robotic movements and resource allocation, ensuring real-time responsiveness to dynamic manufacturing conditions. Additionally, a hybrid AI model combining Long Short-Term Memory (LSTM) networks with Transformer architectures will enable predictive analytics for fault detection and workflow optimization. The AR interface will provide real-time visualization of AI-driven recommendations, guiding human operators through intelligent overlays and haptic feedback mechanisms for seamless human-AI collaboration. The approach aims to develop a scalable, adaptable, and self-learning manufacturing framework that enhances productivity, reduces downtime, and ensures high precision in industrial environments. The proposed AI-AR hybrid system will significantly advance Industry 4.0 and smart manufacturing, bridging the gap between human expertise and machine intelligence through a data-driven, learning-based approach.