Two-Legged Robot Design and System Identification with Deep Learning Iki Bacakli Robot Tasarimi ve Derin Ö?grenme ile Sistem Tanilamasi
34th Signal Processing and Communications Applications Conference, SIU 2026, İstanbul, Türkiye, 7 - 10 Temmuz 2026, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Doi Numarası: 10.1109/siu71813.2026.11636850
- Basıldığı Şehir: İstanbul
- Basıldığı Ülke: Türkiye
- Anahtar Kelimeler: artificial neural networks, deep learning, robotics, system identification
- Yıldız Teknik Üniversitesi Adresli: Evet
Özet
Today, different artificial neural network applications are being used in various robot systems. Especially in the field of legged robots, important advancements are made in recent years. In this study, data obtained from a two-legged robot system developed in our laboratory are processed by artificial neural networks to be used in robot system identification. Results obtained are presented comparatively for this study, in which different activation functions are used and DELU activation function utilization acquired the lowest test set error rate.