Matthias Schöpfer

A Probabilistic Approach to Tactile Shape Reconstruction

In this paper, we present a probabilistic spatial approach to build compact 3-D representations of unknown objects probed by tactile sensors. Our approach exploits the high frame rates provided by modern tactile sensors and utilizes Kalman filters to build a probabilistic model of the contact point cloud that is efficiently stored in a kd-tree. The quality of generated shape representations is compared with a naive averaging approach, and we show that our method provides superior accuracy. We also evaluate the feasibility of object classification combining the generated object representations, together with the iterative closest point algorithm.

Dynamic tactile sensing for object identification

We propose a neural architecture for the recognition of objects by haptics. We demonstrate its performance for a set of household objects and toys using a low cost 2D pressure sensor of coarse resolution, which is moved by a robot arm guided by contact points. The approach transfers the well known view-based method from computer vision to the domain of tactile sensing. However, in contrast to computer vision, not static frames but entire time series of 2D pressure profiles are evaluated.

Identifying Relevant Tactile Features for Object Identification

Tactile sensing arrays for robotic applications become more and more popular these days. This allows us to equip robots with sensing abilities similar to those of our human skin. This article presents an approach to tactile-based recognition of objects and evaluates the utility of various feature extractors for tactile processing.

Open source real-time control software for the Kuka light weight robot

The Kuka light weight robot offers unique features to researchers. Besides its 7 Degrees of Freedom (DOF), also torque sensing in every joint and a variety of compliance modes make the robot a good choice for robotic research. Unfortunately the interface to control the robot externally has its restrictions. In this paper, we present an open source solution (OpenKC) that will allow the control of the robot externally using a simple set of routines that can easily be integrated in existing software. All features and modes of the Kuka light weight robot can be used and triggered externally. Simultaneous control of several robots is explicitly supported. The software has proven its use in several applications.

Automated Door Detection with a 3D-Sensor

Service robots share the living space of humans. Thus, they should have a similar concept of the environment without having everything labeled beforehand. The detection of closed doors is challenging because they appear with different materials, designs and can even include glass inlays. At the same time their detection is vital in any kind of navigation tasks in domestic environments. A typical 2D object recognition algorithm may not be able to handle the large optical variety of doors. Improvements of low-cost infrared 3D-sensors enable robots to perceive their environment as spatial structure. Therefore we propose a novel door detection algorithm that employs basic structural knowledge about doors and enables to extract parts of doors from point clouds based on constraint region growing. These parts get weighted with Gaussian probabilities and are combined to create an overall probability measure. To show the validity of our approach, a realistic dataset of different doors from different angles and distances was acquired.

A High-Speed Tactile Sensor for Slip Detection

Dexterous grasping and manipulation of objects with robot hands requires the ability to monitor contact locations in real-time and with good spatial resolution in order to close the control loop required for object and contact trajectory generation. The ability to recognize incipient slippage will allow for autonomous grasp force adaption – a major prerequisite to handle objects of unknown weight.

Grasping Objects of Unknown Geometry with Tactile Feedback

Service robots operating in unconstrained environments like our homes need to be able to grasp and manipulate all kinds of objects, even if they are not known in advance.While many common approaches for grasping rely on detailed geometric models of the objects to perform a grasp optimization process in simulated scenes, we propose a model-free approach which relies on tactile feedback and coarse shape estimation from vision. Tactile sensor arrays continuously shrink in size and become available for small-size finger tips these days, allowing us to consider more natural approaches to grasping. Evaluating this grasping approach on two different hands, the anthropomorphic Shadow Dexterous Hand and the three-fingered Schunk Dexterous Hand, we prove the feasibility and portability of our method.