Düşük Maliyetli Robotik Kol Aracılığıyla Üretken Yapay Zeka Görüntülerinin Çizilmesi

Öz

Bu çalışma, görsel içerik oluşturmak ve fiziksel olarak işlemek için üretken yapay zekayı robotik kolla birleştiren yapay zeka odaklı bir sistem önermektedir. Sistem, en son metinden görüntüye modellerini kullanarak, önce doğal dil komutlarına dayalı yüksek çözünürlüklü görüntüler oluşturur. Daha sonra bu görüntüler işlenir ve robotik yürütme için uygun vektör yollarına çevrilir. Üretilen içeriği fiziksel bir ortama çizmek için özellikle BrachioGraph platformu olan bir robotik kol kullanılır ve dijital üretim ile somut gerçekleştirme arasındaki boşluğu etkili bir şekilde kapatır. Sistem mimarisi, derin öğrenmeye dayalı görüntü sentezini, yol optimizasyon algoritmalarını ve hassas robotik kontrolü birleştirir. Çalışma, Raspberry Pi ve basit ekipmanlarla oluşturulduğu için titreşime neden olan belirli optimizasyon hatalarına sahip olsa da, yapay zeka teknolojileriyle sanatsal çalışmaların yaratılabileceğini gösteren bir prototip çalışmadır. Bu disiplinlerarası yaklaşım, yapay zeka tarafından üretilen yaratıcılığın robotik aracılığıyla fiziksel olarak tezahür ettiği, otomatik içerik oluşturma, insan-makine işbirliği ve yapay zeka odaklı sanatsal ifadede yeni olanaklar açan yeni bir akışı göstermektedir.

Anahtar Kelimeler

• BrachioGraph
• Üretken Yapay Zeka
• Görüntü Çizimi
• İnsan-Makine Etkileşimi

Plotting of Generative AI-Images via Low-Cost Robotic Arm

Öz

This study proposes an AI-driven system that integrates generative artificial intelligence with robotic arm to create and physically render visual content. Utilizing state-of-the-art text-to-image models, the system first generates high-resolution images based on natural language prompts. These images are then processed and translated into vector paths suitable for robotic execution. A robotic arm, specifically the BrachioGraph platform, is employed to draw the generated content on a physical medium, effectively bridging the gap between digital generation and tangible realization. The system architecture combines deep learning-based image synthesis, path optimization algorithms, and precise robotic control. Although the study has certain optimization errors that cause vibration because it was created with Raspberry Pi and simple equipments, it is a prototype study that shows that artistic studies can be created by AI technologies.This interdisciplinary approach demonstrates a novel pipeline where AI-generated creativity is physically manifested through robotics, opening new possibilities in automated content creation, human-machine collaboration, and AI-driven artistic expression.

Anahtar Kelimeler

• BrachioGraph
• Generative AI
• Robotic Arm
• Image Plotting
• Human-Machine Interaction

Kaynakça

1. [1] K. Dheenadhayalan et al., “AI Art Generators: Human Creativity Vs Artificial Intelligence,” in Proc. 2024 Int. Conf. Power, Energy, Control and Transmission Systems (ICPECTS), IEEE, 2024.
2. [2] S. Göring, R. R. Ramachandra Rao, R. Merten, and A. Raake, “Appeal and quality assessment for AI-generated images,” in Proc. 15th Int. Conf. Quality of Multimedia Experience (QoMEX), Ghent, Belgium, 2023, pp. 115–118.
3. [3] S. Song, J. Song, J. Lee, Y. Kang, and H. Moon, “Exploring the potential of novel image-to-text generators as prompt engineers for CivitAI models,” in Proc. 16th IIAI Int. Congr. Advanced Applied Informatics (IIAI-AAI), Takamatsu, Japan, 2024, pp. 626–631.
4. [4] O. Zambrano and B. Senouci, “Image classification improvement: Text-to-image AI for synthetic dataset approach,” in Proc. 49th Euromicro Conf. Software Eng. and Advanced Applications (SEAA), Durres, Albania, 2023, pp. 74–77.
5. [5] L. Fu, “Research on the construction system of computer big data in the community integrated garbage classification AI image recognition platform,” in Proc. IEEE Conf. Telecommunications, Optics and Computer Science (TOCS), Dalian, China, 2022, pp. 643–647.
6. [6] R. K. Megalingam, S. Boddupalli, and K. G. S. Apuroop, “Robotic arm control through mimicking of miniature robotic arm,” in Proc. 4th Int. Conf. Advanced Computing and Communication Systems (ICACCS), Coimbatore, India, 2017, pp. 1–7.
7. [7] B. Díaz, N. Pacheco, and L. Vinces, “Integration of a robotic arm Lynxmotion to a Robotino Festo through a Raspberry Pi 4,” in Proc. IEEE Int. Conf. Automation / XXV Congr. Chilean Assoc. Automatic Control (ICA-ACCA), Curicó, Chile, 2022, pp. 1–5.
8. [8] A. Bidgoli, M. L. De Guevara, C. Hsiung, J. Oh, and E. Kang, “Artistic style in robotic painting; a machine learning approach to learning brushstroke from human artists,” in Proc. 29th IEEE Int. Conf. Robot and Human Interactive Communication (RO-MAN), Naples, Italy, 2020, pp. 412–418.

9. [9] K. McAleer, BrachioGraph, Kev’s Robots, Jul. 26, 2024. [Online]. Available: https://www.kevsrobots.com/blog/brachiograph.html [Accessed: Jan. 8, 2025].
10. [10] SG90 Datasheet. [Online]. Available: http://www.ee.ic.ac.uk/pcheung/teaching/DE1_EE/stores/sg90_datasheet.pdf [Accessed: Mar. 7, 2025].
11. [11] Stability AI, “DreamStudio API,” Stability AI Developer Platform, Accessed: Apr 22, 2025. [Online]. Available: https://platform.stability.ai/docs/api-reference
12. [12] Stable Diffusion Art - Models. [Online]. Available: https://stable-diffusion-art.com/models/ [Accessed: Feb 05, 2025].
13. [13] S. Jamal, H. Wimmer, and C. M. Rebman Jr., “Perception and evaluation of text-to-image generative AI models: A comparative study of DALL-E, Google Imagen, GROK, and Stable Diffusion,” Issues in Information Systems, vol. 25, no. 2, pp. 277–292, 2024.
14. [14] C. Zhang, C. Zhang, M. Zhang, and I. S. Kweon, “Text-to-image diffusion models in generative AI: A survey,” Journal of LaTeX Class Files, vol. 14, no. 8, pp. 1–20, Aug. 2015.
15. [15] M. Kang, J.-Y. Zhu, R. Zhang, J. Park, E. Shechtman, S. Paris, and T. Park, “Scaling up GANs for text-to-image synthesis,” arXiv preprint arXiv:2303.05511, Mar. 2023. [Online]. Available: https://arxiv.org/abs/2303.05511
16. [16] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Image quality assessment: From error visibility to structural similarity,” IEEE Transactions on Image Processing, vol. 13, no. 4, pp. 600–612, Apr. 2004.
17. [17] A. Sciarra, S. Chatterjee, M. Dünnwald, G. Placidi, A. Nürnberger, O. Speck, and S. Oeltze-Jafra, “Automated SSIM Regression for Detection and Quantification of Motion Artefacts in Brain MR Images,” arXiv preprint arXiv:2206.06725, Jun. 2022. [Online]. Available: https://arxiv.org/abs/2206.06725
18. [18] A. Hore and D. Ziou, “Image quality metrics: PSNR vs. SSIM,” in Proc. 20th International Conference on Pattern Recognition (ICPR), Istanbul, Turkey, Aug. 2010, pp. 2366–2369.
19. [19] G. Prieto Renieblas, A. Turrero Nogués, A. Muñoz González, N. Gómez-Leon, and E. Guibelalde del Castillo, “Structural similarity index family for image quality assessment in radiological images,” Journal of Medical Imaging, vol. 4, no. 3, p. 035501, Jul. 2017.
20. [20] U. Sara, M. Akter, and M. S. Uddin, “Image Quality Assessment through FSIM, SSIM, MSE and PSNR—A Comparative Study,” Journal of Computer and Communications, vol. 7, no. 3, pp. 8–18, Mar. 2019. https://doi.org/10.4236/jcc.2019.73002
21. [21] T. Bayrak et al., “Raspberry Pi based Object Detection and Drawing,” in Proc. Global Conf. on Engineering Research (GLOBCER’21), Balıkesir, Turkey, 2021, pp. 194–201.