Mario Richtsfeld

Point Cloud Segmentation Based on Radial Reflection

This paper introduces a novel 3D segmentation algorithm, which works directly on point clouds to address the problem of partitioning a 3D object into useful sub-parts. In the last few decades, many different algorithms have been proposed in this growing field, but most of them are only working on complete meshes. However, in robotics, computer graphics, or other fields it is not always possible to work directly on a mesh. Experimental evaluations of a number of complex objects demonstrate the robustness and the efficiency of the proposed algorithm and the results prove that it compares well with a number of state-of-the-art 3D object segmentation algorithms.

Grasping unknown objects based on 2½D range data

The problem of grasping novel objects in a fully automatic way has gained increasing importance. In this work we consider the problem of grasping novel objects with the help of a laser range scanner. This includes autonomous object detection and grasp motion planning. The used system consists of a fixed working station equipped with a laser range scanner, a seven degrees of freedom manipulator and a hand prosthesis as gripper. We present a method for segmentation of a 2frac12D point cloud into parts, assembly of parts into objects and calculation of grasping points, which works for cylindrical objects and arbitrary objects. We successfully demonstrate this approach by grasping a variety of different shapes and present a step towards full automation.

Robotic Grasping of Unknown Objects

This work describes the development of a novel vision-based grasping system for unknown objects based on laser range and stereo data. The work presented here is based on 2.5D point clouds, where every object is scanned from the same view point of the laser range and camera position. We tested our grasping point detection algorithm separately on laser range and single stereo images with the goal to show that both procedures have their own advantages and that combining the point clouds reaches better results than the single modalities. The presented algorithm automatically filters, smoothes and segments a 2.5D point cloud, calculates grasping points, and finds the hand pose to grasp the desired object.

Optical Seam Following for Automated Robot Sewing

Nowadays, robust and light-weight parts used in the automobile and aeronautics industry are made of carbon fibres. To increase the mechanical toughness of the parts the carbon fibres are stitched in the preforming process using a sewing robot. However, current systems miss high flexibility and rely on manual programming of each part. The main target of this work is to develop an automatic system that autonomously sets the structure strengthening seams. Therefore, a rapid and flexible following of the carbon textile edges is required. Due to the black and reflective carbon fibres a laser-stripe sensor is necessary and the processing of the range data is a challenging task. The paper proposes a real time approach where different edge detection methodologies are combined in a voting scheme to increase the edge tracking robustness. The experimental results demonstrate the feasibility of a fully automated, sensor-guided robotic sewing process. The seam can be located to within 0.65mm at a detection rate of 99.3% for individual scans.

Seam Following for Automated Industrial Fiber Mat Stitching

This paper presents a method for automatic seam following of two overlapping carbon fiber mats based on laser scans. We introduce a novel approach that combines one existing and two newly developed edge detection methods in a two out of three voting scheme to obtain high edge tracking robustness. The experimental results demonstrate the feasibility of a fully automated, sensor-guided robotic stitching process. The seam can be located within 1.0 mm at a detection rate of 99.3%.

Real-time edge detection for automated fibre-mat stitching

We present a closed process chain for robotic sewing from CAD design to the component model to realize a lot-size-one production. To achieve a better solidity and durability of two overlapping carbon fibre layers, an automated fibre mat stitching system was developed. At the moment, the stitching path is predefined and inflexible to minor changes. To enable a more flexible autonomous stitching process than given through teach-in programming, sensors are used to calculate the difference between the CAD sewing track and the actually required stitching position. The application of the sensor guided stitching process makes it possible for the robot to handle inaccuracies during inlaying of the fibre mats into the form autonomously. Additionally, possible obstacles and mat-ends are detected to realize a stable stitching process.

Detection of cylindrical objects in tabletop scenes

This paper presents a system with a fixed robot arm and a scanning unit on a table, which is able to detect and grasp given cylindrical objects with cluttered adjacent objects in soft real-time. In the fields of industrial and home robotics, the requirements of complete 3D data, noiselessness, and obstacle-free situations are often not provided. The contribution of this work is a fast and robust method optimised for fitting cylinders in sparse and noisy range data under difficult and changing light conditions recorded from a single view. The improvements focus on the treatment of different objects on the table. The system must distinguish between them, detect, and grasp the given cylindrical object.

Lösungen zur Fertigung kleiner Stückzahlen in vielen Varianten

ZusammenfassungKundenorientierung fordert höchst flexible Fertigung mit immer kleineren Losgrößen. Dieser Artikel stellt Verfahren vor, die durch ein Vermessen der Form mittels Lasersensoren erreichen, dass der Fertigungsprozess jedem festigenden Teil speziell angepasst werden kann. Die Erkennung kleiner Stufen erlaubt, der Naht beim Roboternähen zu folgen. Die exakte Vermessung der Position und Orientierung von Bohrungen ermöglicht eine automatisierte Bohrlochprüfung. Die Erfassung der kompletten Bauteilgeometrie gestattet es, Bauteile in Losgröße eins zu lackieren. Letztendlich wird gezeigt, dass die visuelle Erfassung auch ein flexibles Greifen mit dem Roboter erlaubt. Die Genauigkeit liegt jeweils bei 1:100, z. B. wird das Zentrum des Bohrlochs bis auf 0,3 mm vermessen. Die Ergebnisse zeigen, dass die Berechnungen in Echtzeit erfolgen und der Messort direkt beim Fertigungsort liegt, um bis zu Losgröße eins zu realisieren.SummaryCustomer orientation requests highest flexibility of production processes. This paper presents work that enables adaptation to each part by measuring the object shape using laser scanning. Small height differences are detected to enable robot sewing. Exact pose estimation of bore holes allows bore inspection. Identifying elementary part geometries is the way to spray paint unknown convex parts. Finally the image processing methods are used to extract object shape for robotic grasping. The achievable accuracy is better than 1:100, depending on object size, and reaches, e.g., 0.3 mm in bore hole estimation. The results indicate that processing is feasible in real time and that the laser scanning can be placed in the production line.