Generative AI-Assisted Robot Task Planning in Industrial Applications

Abstract

The rapid evolution of Industry 4.0 has accelerated the adoption of autonomous robotic systems in intelligent manufacturing. Traditional robot task planning methods are limited in dynamic production environments due to their dependence on rule-based programming and deterministic algorithms. In contrast, Generative Artificial Intelligence (GenAI), including Large Language Models (LLMs) and multimodal foundation models, enables robots to understand natural language, perform contextual reasoning, generate adaptive task plans, and respond intelligently to changing industrial conditions. This paper presents a Generative AI-based framework for robotic task planning that integrates multimodal perception, semantic reasoning, LLM-based planning, reinforcement learning, and execution monitoring. The framework enhances planning accuracy, adaptability, collaboration, and operational efficiency while reducing execution errors and planning complexity. It is applicable to smart assembly, automated inspection, warehouse logistics, predictive maintenance, collaborative robotics, and flexible manufacturing. The study concludes that Generative AI offers a promising foundation for scalable, adaptive, and human-centric industrial automation, with future research directed toward continual learning, explainable AI, federated robotic intelligence, edge AI, and trustworthy autonomous decision-making.