Refine
Year of publication
Document Type
- Conference Proceeding (46) (remove)
Keywords
- LiDAR (3)
- Agile Lehre (2)
- Gazebo (2)
- Inverted Classroom (2)
- Power Consumption (2)
- Robot operating system (ROS) (2)
- Scrum (2)
- Simulation and Modeling (2)
- biogas (2)
- mobile field laboratory (2)
Institute
- Fakultät IuI (46) (remove)
The usage of high-level synthesis (HLS) tools for FPGAs has increased significantly over the last years since they matured and allow software programmers to take advantage of reconfigurable hardware technology.
Most HLS tools employ methods to optimize for loops, e. g. by unrolling or pipelining them. But there is hardly any work on the optimization of while loops. This comes at no surprise since most while loops have loop-carried dependences involving the loop condition which result in large recurrence cycles in the dataflow graphs. Therefore typical while loops cannot be parallelized or pipelined.
We propose a novel transformation which allows to optimize while loops nested within a for loop. By interchanging the two loops, it is possible to pipeline (and thereby parallelize) the inner loop, resulting in a reduced execution time. We present two case studies on different hardware platforms and show the speedup factors - compared to a host processor and to an unoptimized hardware implementation - achieved by our while loop optimization method.
Artificial intelligence (AI) promises transformative impacts on society, industry, and agriculture, while being heavily reliant on diverse, quality data. The resource-intensive "data
problem" has initialized a shift to synthetic data. One downside of synthetic data is known as the "reality gap", a lack of realism. Hybrid data, combining synthetic and real data, addresses this. The paper examines terminological inconsistencies and proposes a unified taxonomy for real, synthetic, augmented, and hybrid data. It aims to enhance AI training datasets in smart agriculture, addressing the challenges in the agricultural data landscape. Utilizing hybrid data in AI models offers improved prediction performance and adaptability.
This paper presents an optimized algorithm for estimating static and dynamic gait parameters. We use a marker- and contact-less motion capture system that identifies 20 joints of a person walking along a corridor.
Based on the proposed gait cycle detection basic metrics as walking frequency, step/stride length, and support phases are estimated automatically. Applying a rigid body model, we are capable to calculate static and dynamic gait stability metrics. We conclude with initial results of a clinical study evaluating orthopaedic technical support.
he development of context-aware applications is a difficult and error-prone task. The dynamics of the environmental context combined with the complexity of the applications poses a vast number of possibilities for mistakes during the creation of new applications. Therefore it is important to test applications before they are deployed in a life system. For this reason, this paper proposes a testing tool, which will allow for automatic generation of various test cases from application description documents. Semantic annotations are used to create specific test data for context-aware applications. A test case reduction methodology based on test case diversity investigations ensures scalability of the proposed automated testing approach.
Artificial intelligence (AI) and human-machine interaction (HMI) are two keywords that usually do not fit embedded applications. Within the steps needed before applying AI to solve a specific task, HMI is usually missing during the AI architecture design and the training of an AI model. The human-in-the-loop concept is prevalent in all other steps of developing AI, from data analysis via data selection and cleaning to performance evaluation. During AI architecture design, HMI can immediately highlight unproductive layers of the architecture so that lightweight network architecture for embedded applications can be created easily. We show that by using this HMI, users can instantly distinguish which AI architecture should be trained and evaluated first since a high accuracy on the task could be expected. This approach reduces the resources needed for AI development by avoiding training and evaluating AI architectures with unproductive layers and leads to lightweight AI architectures. These resulting lightweight AI architectures will enable HMI while running the AI on an edge device. By enabling HMI during an AI uses inference, we will introduce the AI-in-the-loop concept that combines AI's and humans' strengths. In our AI-in-the-loop approach, the AI remains the working horse and primarily solves the task. If the AI is unsure whether its inference solves the task correctly, it asks the user to use an appropriate HMI. Consequently, AI will become available in many applications soon since HMI will make AI more reliable and explainable.
Innovationen sind die stärksten Gestaltungsfaktoren für eine neue vielversprechende Zukunft, da sie die wichtigsten Treiber für Wachstum und Ertrag in unserer Wirtschaft sind. Die aktuelle Zeitenwende zeigt uns sehr deutlich, dass wir ohne Innovationen bzw. Veränderungen und Anpassungen kaum noch wettbewerbsfähig bleiben, sowohl als Nation bzw. als Gesellschaft und insbesondere als Unternehmen.
Die hohe Dynamik und Komplexität der wirtschaftlichen und sozialen Prozesse setzt neue Maßstäbe an die Innovationsstrategien von Institutionen und Unternehmen.
Neue Technologien, neue Märkte, neues Kundenverhalten und der stetige Wandel sowohl in der Arbeitswelt als auch in unserem gesellschaftlichen Umfeld, wie z.B. die Digitalisierung, zeigen uns, dass allein eine Produktinnovation als solche heute nicht mehr ausreicht. Unter den genannten Randbedingungen müssen Innovationen auch in der Gestaltung von Geschäftsprozessen und Realisierung der "Work-Life-Balance" neu erdacht bzw. überprüft werden.
Der Vorsprung innovativer Produkte im viralen Wettbewerb ist oft nur kurz. Ein ganzheitliches Innovationsmanagement hat alle Bereiche des Unternehmens einzubeziehen und führt zu neuen Geschäftsmodellen, die etablierte Geschäftspraktiken verdrängen, ebenso tauchen durch neue Technologien in immer stärkerem Maße neue Anbieter auf, die die Spielregeln in den Märkten verändern.
Der 1. Deutsche Innovations-Kongress will Impulse setzen, Best-Practice-Modelle als Vorbilder anbieten und im Austausch zwischen den Referent*innen und den Teilnehmer*innen neue Wege bzw. Perspektiven eröffnen.
Wir freuen uns auf alle Teilnehmer*innen und den Erfahrungsaustausch, um aktuelle und nachhaltige Innovations-Impulse zu setzen und neue Wege erfolgversprechende Wege zu beschreiben, womit die bereits fruchtbaren Kooperationen zwischen Wirtschaft und Wissenschaft im Großraum Osnabrück noch weiter belebt werden soll.
Aktuell tragen auch 8 Studierendengruppen des Masterstudiengangs "Entwicklung und Produktion" der Hochschule Osnabrück in der Fakultät I u. I im Rahmen des Moduls "Innovationsmanagement" in Kooperation mit Unternehmen aus der Region durch die Entwicklung neuer innovativer Produkte zum Erfolg des Kongresses bei. Die Zwischenergebnisse dazu werden in einer Poster-Ausstellung präsentiert. Die Innovationsprojekte werden unter der Leitung von Prof. Dr. Jens Schäfer durchgeführt.
Bamboo is an environmentally friendly alternative to conventional materials in mechanical engineering such as steel or aluminium. Bamboo is the fastest growing plant in the world. Instead of releasing CO2 during the manufacturing process, bamboo absorbs CO2 as it grows.
In addition to the sustainability aspect, bamboo tubes also offer excellent properties as a lightweight construction material, which have been optimised through evolution. Bamboo tubes have high strength and stiffness at low weight when used as tension-compression bars or bending beams. Bamboo has strong, high-density fibres at the boundary area, where bending stresses are greatest. Towards the inside, where the stresses are lower, the bamboo becomes porous to optimise weight. This, together with knots arranged in regular intervals, counteracts buckling.
In mobile applications such as cars and bicycles, lightweight construction is sought for energy efficiency reasons. Because of its excellent lightweight properties, the project investigated whether bamboo could be used in mobile, automotive or agricultural engineering. For example, a bamboo bicycle frame has been developed with the aim to be as light as possible. There are bamboo bicycles on the market, but they can only be made one at a time by hand. The bamboo tubes are joined together and functional elements such as the bottom bracket and headset are integrated by wrapping them in resin-impregnated natural or carbon fibres. This makes the joints very heavy. A different approach is taken here: the bamboo tubes are drilled out slightly to achieve a defined internal diameter, and then short aluminium tubes are glued into the bamboo canes from the inside. To prevent the cane from breaking in the circumferential direction, i.e. perpendicular to the fibre direction, the bamboo tubes are wrapped in a thin layer of natural or carbon fibre impregnated with synthetic resin. The aluminium tubes and functional elements are welded or soldered together beforehand.
The design of the bicycle frame, i.e. the dimensioning of the bamboo tubes and joints, was based on extensive bending and tensile tests to determine the strength properties of the natural material bamboo. The bonding between the bamboo cane and the aluminium tube was also investigated experimentally. Finally, several prototype bicycle frames were made and tested for durability according to DIN-EN-14764. The frames passed the tests.
The result is a bamboo bicycle that is manufactured with standardised connectors and joints. The assembly concept developed allows both fully automated and semi-automated series production of bamboo bicycles.
With the increasing size and complexity of embedded systems, the impact of software on energy consumption is becoming more important. Previous research focused mainly on energy optimization at the hardware level. However, little research has been carried out regarding energy optimization at the software design level. This paper focuses on the software design level and addresses the gap between software and hardware design for embedded systems. This is achieved by proposing a framework for software design patterns, which takes aspects of power consumption and time behavior of the hardware level into account. We evaluate the expressiveness of the framework by applying it to well-known and novel design patterns. Furthermore, we introduce a dimensionless numerical efficiency factor to make possible energy savings quantifiable.
Due to the resource-constrained nature of embedded systems, it is crucial to support the estimation of their power consumption as early in the development process as possible. Non-functional requirements based on power consumption directly impact the software design, e.g., watt-hour thresholds and expected lifetimes based on battery capacities. Even if software affects hardware behavior directly, these types of requirements are often overlooked by software developers because they are commonly associated with the hardware layer. Modern trends in software engineering such as Model-Driven Development (MDD) can be used in embedded software development to evaluate power consumption-based requirements in early design phases. However, power consumption aspects are currently not sufficiently considered in MDD approaches. In this paper, we present a model-driven approach using Unified Modeling Language profile extensions to model hardware components and their power characteristics. Software m odels are combined with hardware models to achieve a system-wide estimation, including peripheral devices, and to make the power-related impact in early design stages visible. By deriving energy profiles, we provide software developers with valuable feedback, which may be used to identify energy bugs and evaluate power consumption-related requirements. To demonstrate the potential of our approach, we use a sensor node example to evaluate our concept and to identify its energy bugs.
Optimised Nutrient Recovery from Biogas Digestate by Solid/Liquid Separation and Membrane Treatment
(2019)
Anaerobic digestion products of agricultural biogas plants are characterised by high nitrogen, phosphorus, and potassium content. In three scale-up steps, a membrane based digestate treatment process of solid-liquid-separation, ultrafiltration, and reverse osmosis for nutrient recovery was investigated. Lab-scale trials delivered a very good understanding of fluid properties and subsequent ultrafiltration performance, which is the limiting process step in terms of energy demand and operation costs. In semi-technical experiments, optimisation, and design parameters were developed, which were subsequently applied to pilot-scale tests at two full-scale biogas plants. The process optimisation resulted in 50 % energy reduction of the ultrafiltration step. About 36 % of the sludge volume was recovered as dischargeable water, 20 % as solid N/P-fertiliser, and 44 % as liquid N/K-fertiliser.
Currently, soil nutrient analysis involves two separate processes for soil sampling and nutrient analysis: 1. field soil sampling and 2. laboratory analysis. These two - separate - main work processes are combined and conceptualised for a mobile field laboratory so that soil sampling and analysis can be carried out simultaneously in the field. The module-based field laboratory "soil2data" can carry out these two main work processes in parallel and consists of 5 different task-specific modules that build on each other: app2field, field2soil, app2liquid, liquid2data and data2app. The individual modules were designed and built for the sub-process steps and adapted to the special features of the mobile field laboratory "soil2data". The biggest advantage is that the analysis results are available immediately, and a fertiliser recommendation can be generated instantly. For further analyses, the results are stored in the data cloud. The soil material remains in the field. In the ongoing project "Prototypes4soil2data", the mobile field laboratory soil2data is being further developed into a prototype with a modular structure.
This article proposes the concept of a simulation framework for environmental sensors with multilevel abstraction in agricultural scenarios. The implementation case study is a simulation of a grain-harvesting scenario enabled by LiDAR sensors. Environmental sensor models as well as kinematics and dynamic behavior of machines are based on the robotics simulator Gazebo. Models for powertrain, machine process aggregates and peripheral simulation components are implemented with the help of MATLAB/ Simulink and with the robotics middleware Robot Operating System (ROS). This article deals with the general concept of a multilevel simulation framework and in particular with sensor and environmental modeling.
Simulation von Laserscannern in Pflanzenbeständen für die Entwicklung umfeldbasierter Funktionen
(2018)
Es werden drei Modellierungsansätze zur Simulation von Laserscannern in Pflanzenbeständen für die Entwicklung umfeldbasierter Fahrzeugfunktionen beschrieben. Das Sensorsignal der Distanzmessung wird dabei anhand realer Messwerte oder phänomenologisch und auf der Basis empirisch ermittelter Kennwerte in Abhängigkeit von objekt- und sensorspezifischen Einflussfaktoren abgebildet. Basierend auf den Methoden zur Simulation von Distanzmesssystemen der Open Source Simulationsumgebung Gazebo wurden die Modellierungsansätze als spezifische Sensor- und Umfeldmodelle implementiert. Die Modelle wurden insbesondere für den Einsatz an mobilen landwirtschaftlichen Arbeitsmaschinen und für die Anwendung in der Getreideernte optimiert.
This paper describes the development and test of a novel LiDAR based combine harvester steering system using a harvest scenario and sensor point cloud simulation together with an established simulation toolchain for embedded software development. For a realistic sensor behavior simulation, considering the harvesting environment and the sensor mounting position, a phenomenological approach was chosen to build a multilayer LiDAR model at system level in Gazebo and ROS. A software-in-the-loop simulation of the mechatronic steering system was assembled by interfacing the commercial AppBase framework for point cloud processing and feature detection algorithms together with a machine model and control functions implemented in MATLAB/ Simulink. A test of ECUs in a hardware-in-the-loop simulation and as well as HMI elements in a driver-in-the-loop simulation was achieved by using CAN hardware interfaces and a CANoe based restbus simulation.
Smart city applications in the Big Data era require not only techniques dedicated to dynamicity handling, but also the ability to take into account contextual information, user preferences and requirements, and real-time events to provide optimal solutions and automatic configuration for the end user. In this paper, we present a specific functionality in the design and implementation of a declarative decision support component that exploits contextual information, user preferences and requirements to automatically provide optimal configurations of smart city applications. The key property of user-centricity of our approach is achieved by enabling users to declaratively specify constraints and preferences on the solutions provided by the smart city application through the Decision Support component, and automatically map these constraints and preferences to provide optimal responses targeting user needs. We showcase the effectiveness and flexibility of our solution in two real usecase scenarios: a multimodal travel planner and a mobile parking application. All the components and algorithms described in this paper have been defined and implemented as part of the Smart City Framework CityPulse.
Management of agricultural processes is often troubled by disconnections and data transfer failures. Limited cellular network coverage may prevent information exchange between mobile process participants.
The research projects KOMOBAR and ISOCom designed, implemented und field-tested a delay tolerant platform for robust communication in rural areas and challenging environments. An adaptable combination of infrastructure-based cellular networks and infrastructure-free multihop ad hoc communication (WLAN) leads to a variety of new communication opportunities. Temporal storage and forwarding of data on mobile farm machinery as well as dynamic platform configurations during process runtime strongly enhance reliability and robustness of data transfers.
Process modeling languages help to define and execute processes and workflows. The Business Process Model and Notation (BPMN) 2.0 is used for business processes in commercial areas such as banks, shops, production and supply industry. Due to its flexible notation, BPMN is increasingly being used in non-traditional business process domains like Internet of Things (IoT) and agriculture. However, BPMN does not fit well to scenarios taking place in environments featuring limited, delayed, intermittent or broken connectivity. Communication just exists for BPMN - characteristics of message transfers, their priorities and connectivity parameters are not part of the model. No backup mechanism for communication issues exists, resulting in error-prone and failing processes. This paper introduces resilient BPMN (rBPMN), a valid BPMN extension for process modeling in unreliable communication environments. The meta model addition of opportunistic message flows with Quality of Service (QoS) parameters and connectivity characteristics allows to verify and enhance process robustness at design time. Modeling of explicit or implicit, decision-based alternatives ensures optimal process operation even when connectivity issues occur. In case of no connectivity, locally moved functionality guarantees stable process operation. Evaluation using an agricultural slurry application showed significant robustness enhancements and prevented process failures due to communication issues.
Ein modulares Framework zur Modellierung, Konfiguration und Regelung von kooperativen Agrarprozessen
(2016)
Die Komplexität vieler Agrarprozesse nimmt aufgrund von technischem Fortschritt, steigenden rechtlichen Anforderungen und Nachweispflichten beständig zu. Prozessketten werden in Kooperation verschiedener Akteure (Landwirt, Lohnunternehmer, Dienstleister, digitaler Vermittler, Behörde) gemeinsam bearbeitet, dokumentiert und geprüft. Ein ökonomisch und ökologisch ressourceneffizientes Management der Prozessausführung stellt eine Herausforderung für alle Akteure dar. Dynamische Prozessveränderungen führen vielfach zu manuellen Eingriffen in die Prozessregelung, die kostenintensive Verzögerungen verursachen. Das Forschungsvorhaben OPeRAte entwirft und evaluiert neu gestaltete Konzepte und Mechanismen zur durchgehenden Organisation und Regelung kooperativer Agrarprozesse. Es werden konfigurierbare und wiederverwendbare Module identifiziert, die sich an Prozessparameter anpassen und in artverwandten Prozessen erneut verwenden lassen. Das OPeRAte-Framework ermöglicht die Zusammenführung aller beteiligten Akteure und Ressourcen (Maschinen, Sensoren, Aktoren, Endgeräte, Server, Daten, etc.) über offene Schnittstellen. Prozessinhaber sollen durch autonome Prozesskonfigurationen und -adaptionen entlastet und durch Visualisierungen zu effizienten Entscheidungen befähigt werden. Die Konzepte dieses Beitrags dienen als Diskussionsgrundlage zur Formulierung von flexiblen und erweiterbaren Lösungsstrategien für die Landtechnik.
For Delay-Tolerant Networks (DTNs) many routing algorithms have been suggested. However, their performance depends heavily on the applied scenario. Especially heterogeneous scenarios featuring known and unknown node movements as well as different kinds of data lead to either poor delivery ratios or exhausted network resources.
To overcome these problems this paper introduces Data-Driven Routing for DTNs. Data is categorized according to its requirements into priority queues. Each queue applies an appropriate DTN routing algorithm that fits the data requirements best. Simulation results show that Data-Driven Routing allows high delivery ratios for time-critical data while saving network resources during the transfer of less time-critical data at the same time.