Accelerating materials research using AI technologies requires high-quality, reliable data across a wide range of measurement conditions.Powder X-ray diffraction (PXRD) is crucial for analyzing crystal structures and quantifying phase compositions in materials science. However, PXRD reliability heavily depends on accurate sample preparation and data analysis, including precise measurements over a broad angular range, especially at low angles. Recent advancements in AI-driven PXRD data analysis have improved accuracy and efficiency, shifting the bottleneck and reproducibility issues to manual sample preparation and measurement processes. To address these challenges, we developed an autonomous robotic experimentation (ARE) system for PXRD that integrates sample preparation, measurement, and data analysis into a single automated workflow. Our system achieves high precision and reproducibility in sample preparation, enabling quantitative phase analysis with only one-hundredth of the conventional sample quantity (reducing from 300 mg to 3 mg) while maintaining a standard deviation below 1 %. By combining robotic precision with machine learning-based data analysis, our approach enhances reproducibility and enables more efficient materials discovery compared to traditional manual methods.