Grasp Stability from Vision and Touch
Poster (konferens), 2012
We study the exploitation of tactile and visual feedback to predict the stability of a grasp before attempting to lift an object and thus to prevent failures. Our robot learns an empirical representation of stable and unstable grasps by exploring given grasping configurations.
grasping and manipulation, grasp stability, Kernel Logistic Regression
Författare
Yasemin Bekiroglu
Kungliga Tekniska Högskolan (KTH)
Renaud Detry
Kungliga Tekniska Högskolan (KTH)
Danica Kragic
Kungliga Tekniska Högskolan (KTH)
IEEE IROS 2012 Workshop: Advances in Tactile Sensing and Touch-based Human Robot Interaction
Vilamoura, ,
Vilamoura, ,
Ämneskategorier
Robotteknik och automation
Datorseende och robotik (autonoma system)