Grasping and Manipulation of Unknown Objects Based on Visual and Tactile Feedback Robert Haschke Abstract The sense of touch allows humans and higher animals to perform coordinated and efficient interactions within their environment. Recently, tactile sen- sor arrays providing high force, spatial, and temporal resolution became available for robotics, which allows us to consider new control strategies to exploit this impor- tant and valuable sensory channel for grasping and manipulation tasks. Successful dexterous manipulation strongly depends on tight feedback loops integrating propri- oceptive, visual, and tactile feedback. We introduce a framework for tactile servoing that can realize specific tactile interaction patterns, for example to establish and main- tain contact (grasping) or to explore and manipulate objects. We demonstrate and evaluate the capabilities of the proposed control framework in a series of preliminary experiments employing a 16 × 16 tactile sensor array attached to a Kuka LWR arm as a large fingertip. Keywords Grasping · Tactile servoing · Online motion planning 1 Introduction The sense of touch allows humans to perform coordinated and efficient interactions within their environment. Without the sense of touch, subjects have severe difficulties maintaining a stable grasp or performing a complex action such as lightning matches [1, 2]. Also in robot applications, lacking tactile feedback results in loosing an ini- tially grasped object or failing to robustly carry out manipulation tasks [3]. In recent years, the resolution and sensitivity of tactile sensors only sufficed for basic force feedback during blind grasping [4]. However, tactile sensor arrays providing high spatial and temporal resolution as well as high sensitivity [5, 6] emerged recently, allowing for more advanced control methods involving tactile feedback too. R. Haschke (B ) Cognitive Interaction Technology Excellence Cluster (CITEC), Bielefeld University, Inspiration 1, 33619 Bielefeld, Germany e-mail: rhaschke@techfak.uni-bielefeld.de © Springer International Publishing Switzerland 2015 G. Carbone and F. Gomez-Barvo (eds.), Motion and Operation Planning of Robotic Systems, Mechanisms and Machine Science 29, DOI 10.1007/978-3-319-14705-5_4 91