Maximilian Wagner

How to Make Profit: Exploiting Fluctuating Electricity Prices with Albatross, A Runtime System for Heterogeneous HPC Clusters

The ongoing evolution of the power grid towards a highly dynamic supply system poses challenges as renewables induce new grid characteristics. The volatility of electricity sources leads to a fluctuating electricity price, which even becomes negative when excess supply occurs. Operators of high-performance--computing (HPC) clusters therefore can consider the highly dynamic variations of electricity prices to provide an energy-efficient and economic operation. This paper presents Albatross, a runtime system for heterogeneous HPC clusters. To ensure an energy-efficient and economic processing of HPC workloads, our system exploits heterogeneity at the hardware level and considers dynamic electricity prices. We have implemented Albatross and evaluate it on a heterogeneous HPC cluster in our lab to show how the power demand of the cluster decreases when electricity prices are high (i.e., excess demand at the grid). When electricity prices are low or negative (i.e., excess supply to the grid), Albatross purposefully increases the workload and, thus, power demand of the HPC cluster---to make profit.

Self-calibration method for a robotic based 3D scanning system

This paper describes a method for extrinsic sensor calibration for a 2D laser profile sensor on a robot arm used as a robot based 3D scanning system. In order to establish a relationship between the sensor measurements and the robot positions, the transformation between the robot flange and the sensor reference frame are determined. The developed calibration method is implemented as an automated self-calibration. It is based on the use of a pin as a fixed reference. The robot is aligned to the tip of the pin automatically with certain orientations. The sensor transformation is calculated based on six recorded robot positions. Finally, the accuracy of the calibration is determined with an additional robot position.

Making Profit with ALBATROSS: A Runtime System for Heterogeneous High-Performance-Computing Clusters

1 MOTIVATION AND INTRODUCTION The ongoing evolution of the power grid towards a highly dynamic supply and demand system poses challenges [1] to its operators and subscribers. The dependence on renewable electricity (e.g., wind, solar, and water) induces new grid characteristics: the volatility of electricity sources leads to an interplay of excess supply and demand, which results in fluctuating electricity prices. Thus, the operation of high-performance–computing (HPC) clusters canwork with the fluctuating electricity price to ensure cost effectiveness [2]. This poster abstract presents Albatross [3], a runtime system for heterogeneous HPC clusters. To ensure an energy-efficient and economic processing of HPC workloads, our system exploits heterogeneity at the hardware level and considers dynamic electricity prices. Early results of our Albatross prototype running on a heterogeneous HPC cluster in our lab show how the power demand of the cluster decreases when electricity prices are high (i.e., excess demand at the grid), and how our system purposefully increases the workload and, thus, power demand when electricity prices are low or even negative (i.e., excess supply to the grid)—to make profit.

Accuracy Analysis and Improvement for Cooperative Industrial Robots.

The cooperative working of multiple robots on a common task often requires a high geometric accuracy. If such a system is modeled, many sources of error are present, which can quickly lead to inadequate process results. In order to avoid this, it is important to carry out a calibration in which deviations are determined. Subsequently, the model can be adapted to the actual conditions. In the scope of this work a kinematic calibration method for multi-robot systems is developed and realized with a robot setup consisting of two industrial robot arms. The accuracy of the robot system is significantly improved by the developed approach, which has been proven by experimental investigations.

Cooperative processing with multi-robot systems

Multi-robot systems are increasingly utilized for cooperative industrial robot applications. In the scope of this work, a novel approach for the division of robot applications between multiple robots by handling the tool and the workpiece at the same time is presented. The approach has been tested successfully with a robot setup with two industrial robot arms. It is shown that the processing time can be reduced and the reachability can be improved by the developed approach. In order to compensate complex programming of the cooperative tasks, an automatic programming is also realized. Thus, the programming time can be significantly reduced.

Inline calibration method for robot supported process tasks with high accuracy requirements

Especially in cases of robot supported process tasks with high accuracy requirements, exactly known geometric relations of the process relevant cell components relative to each other is of significant importance. To establish these relations, several methods are known from the state of the art, belonging to the area of calibration. However, compensating all influencing factors individually can be very time consuming. Addressing this subject, a more straightforward inline calibration method has been developed in research activities at FAPS, compensating geometrical errors independent from their origin. Therefore, the central effect of inaccuracies is exploited, namely a geometrical misalignment of the process pattern relative to the workpieces. This misalignment is measured via an optical sensor system. Based on this information transformation relations are established and the robot control program subsequently modified in order to adapt the robot movement to the real geometrical relations. In conclusion, the impact of geometrical deviations is compensated. Experimental investigations show a significant improvement of the process result after applying the calibration method.

Intuitive Programmierung für die Mensch-Roboter-Kollaboration

Kurzfassung Aufgrund des wachsenden Einsatzes von Prozessen, in denen Mensch und Roboter zusammen und ohne Schutzzaun interagieren, wächst die Zahl der zu programmierenden sicheren Roboter in der Industrie. Ziel dieser Arbeit ist es, ein Prinzip der einfachen Programmierung zu zeigen, sodass auch Facharbeiter ohne große Programmierkenntnisse eine Mensch-Roboter-Kollaboration (MRK) erstellen können. Es wird ein Baukastenprinzip vorgestellt, mit dem Mensch-Roboter-Interaktionen durch verschieden zusammengesetzte Bausteine aufgebaut beziehungsweise programmiert werden können. Zum Schluss verdeutlichen zwei Beispiele an einem Leichtbauroboter das ausgeführte Prinzip der Programmierung.

Data Fusion Between a 2D Laser Profile Sensor and a Camera

This paper describes a color extension of a 2D laser profile sensor by extracting the corresponding color from a camera image. For these purpose, we developed a routine for an extrinsic calibration between the profile sensor and the camera. Based on the resulting translation and rotation vectors a belonging pixel can be calculated for each profile point. Consequently, the color for each profile point can be extracted from the image. This approach is used to extend the geometric data of a robotic based 3D scanning system by color data.

Mensch-Roboter-Kollaboration in der Fertigung der Zukunft

Kurzfassung Mensch-Roboter-Kollaboration ist ein wichtiger Baustein für Industrie 4.0. Neben der Sicherheit der beteiligten Menschen ist eine Anpassung der Fertigungsprozesse an die Mensch-Roboter-Kollaboration wesentlich für deren erfolgreichen Einsatz. In diesem Beitrag werden grundsätzliche Strukturen für diese Fertigungsprozesse vorgestellt und anhand eines Beispiels wird ein mögliches Kollaborationskonzept in der Montage aufgezeigt.