Towards Grasp-Oriented Visual Perception for Humanoid Robots

Abstract

A distinct property of robot vision systems is that they are embodied. Visual information is extracted for the purpose of moving in and interacting with the environment. Thus, different types of perception-action cycles need to be implemented and evaluated.In this paper, we study the problem of designing a vision system for the purpose of object grasping in everyday environments. This vision system is firstly targeted at the interaction with the world through recognition and grasping of objects and secondly at being an interface for the reasoning and planning module to the real world. The latter provides the vision system with a certain task that drives it and defines a specific context, i.e. search for or identify a certain object and analyze it for potential later manipulation. We deal with cases of: (i) known objects, (ii) objects similar to already known objects, and (iii) unknown objects. The perception-action cycle is connected to the reasoning system based on the idea of affordances. All three cases are also related to the state of the art and the terminology in the neuroscientific area.

A hierarchical Bayesian framework for multimodal active perception

“…Recent work in active vision by Tsotsos & Shubina (2007) and Bohg et al (2009), the former for target search and the latter for object grasping, contrary to our solution, use an explicit representation for objects to implement active perception. On the other hand, several solutions for target applications similar to ours avoid explicit object representation by resorting to a bottom-up saliency approach such as defined by Itti et al (1998) – examples of these would be Shibata, Vijayakumar, Conradt, & Schaal (2001), Breazeal, Edsinger, Fitzpatrick, & Scassellati (2001) and Dankers, Barnes, & Zelinsky (2007).…”

Tactile and Vision Perception for Intelligent Humanoids

“…In Bohg et al,. [ 150 ] a visual–tactile cooperative control framework was proposed, which processed these two kinds of features separately and added them into the feedback control loop to plan the operation together.…”

How to localize humanoids with a single camera?

“…In order to obtain fully autonomous robots an accurate localization of the robot in the world is much more than desirable. Furthermore, if we can obtain an accurate localization in real-time, we can use the remaining computational resources to perform other important humanoid robotic tasks such as planning (Perrin et al, 2010), 3D object modeling (Foissote et al, 2010) or visual perception (Bohg et al, 2009).…”