Optimization of Robotic Arm Scheduling Based on Catfish Optimization Algorithm and Q-Learning

Abstract In order to improve the intelligence and efficiency of robot scheduling in workshop production lines, the author proposes a robot scheduling optimization based on the fish optimization algorithm and Q-learning. Firstly, conduct a thorough analysis of the characteristics of workshop scheduling problems and construct a mathematical model that includes multiple objectives such as minimum completion time and utilization rate of robotic arms. Based on this model, the author designed and implemented a scheduling optimization method that combines catfish optimization algorithm with Q-learning. The innovation of the research lies in integrating these two algorithms to form an adaptive scheduling strategy. By dynamically adjusting the mutation rate in Q-learning and combining it with the population search capability of the fish algorithm, the adaptive ability and global search efficiency of the scheduling process are improved. The experimental results show that in small-scale problems, the hybrid algorithm reduces the average completion time by 8.5% compared to traditional genetic algorithms; In large-scale complex problems, this advantage is even more significant, achieving a 12.7% improvement. Especially when dealing with dynamic scheduling environments, hybrid algorithms exhibit stronger adaptability and robustness. Conclusion: The results of this study are of great significance for promoting the development of intelligent manufacturing technology, improving production efficiency, and reducing production costs. Not only does it enrich the application research of intelligent optimization algorithms in practical engineering problems in theory, but it also provides practical guidance for the transformation of manufacturing industry towards intelligence and flexibility.