Analysis of the learning results of a reinforcement learning AI that suggests scanning paths in the Selective Laser Thermoregulation method

Key Points

• This research aims to optimize the scanning paths in the Selective Laser Thermoregulation method using reinforcement learning.
• Developed Selective Laser Thermoregulation (SLT) method with high-speed laser scanning.
• Utilized a fiber laser and galvanometer for precise irradiation.
• Implemented reinforcement learning AI with Temporal Difference (TD) method to suggest scanning paths.
• The AI efficiently suggests scanning paths that lead to more uniform heating of the metal sample.
• High temperature uniformity is achieved compared to traditional methods.