Landmark Detection via ANN for a Web Based Autonomous Mobile Robot: Sunar
Yükleniyor...
Dosyalar
Tarih
2006
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
In this study, a landmark detection method was developed for finding or position correction of a web based mobile robot designed and implemented for long term and regular scientific purposes. Colored numeric and alphanumeric character sticker in place of other artificial landmarks appropriate for robot is selected to be landmark for understanding of both human and robot. Statistical analysis of captured and segmented image part is used for feature vector extraction. Statistical properties of histogram, projections and image raw data are selectable components of feature vector. The feature vector is tested by previously trained multilayer perceptron feed forward neural network (ANN). For this aim, online programs required for robotic activities, image processing and neural network processes have been developed on web interface of web-robot. In this program, improved software libraries for SUNAR system are employed. Real time results and robot scenes are monitored online on web portal.
Açıklama
Anahtar Kelimeler
Kaynak
Selçuk Journal of Applied Mathematics
WoS Q Değeri
Scopus Q Değeri
Cilt
7
Sayı
1
Künye
Yılmaz, N., Sağıroğlu, Ş., Bayrak, M., (2006). Landmark Detection via ANN for a Web Based Autonomous Mobile Robot: Sunar. Selçuk Journal of Applied Mathematics, 7(1), 45-61.
