Se­lec­ted pro­jects

Simultaneous development and testing of Cyber Physical Systems (CPS) using the application example of an electrically powered autonomous vehicle

dSPACE, e.GO Mobile and the Institute for Industrial Mathematics launch research project

How can autonomous vehicles with electric drives be developed faster, more cost-effectively and with fewer resources as examples of complex cyber-physical systems? And how can the safety of such vehicles on the road be increased? A team of researchers and developers from dSPACE, e.GO Mobile AG and the Institute for Industrial Mathematics at Paderborn University launched a research project a few weeks ago to answer these complex questions. The project is being sponsored by the state of North Rhine-Westphalia and the EU as part of the lead market competition IKT.NRW. "Simultaneous development and testing of Cyber Physical Systems (CPS) using the application example of an electrically powered autonomous vehicle - SET CPS" is the official title of the project, which will run for 36 months.

In the development of vehicles, trends such as automated driving or the development of alternative drive systems, such as battery-powered vehicles, are leading to a sharp increase in the demands placed on the underlying systems. When developing such vehicles, the aim is to optimise a large number of target variables such as consumption, range and driving comfort and to guarantee the safety of the system. Researchers and developers in the SET CPS project are now looking for new approaches to make the development processes reliable and economical for manufacturers and suppliers and to meet development deadlines.

The project therefore aims to develop intelligent, simulation-based processes that improve and systematise the development and testing process of complex vehicles and increase the degree of automation. To this end, design and testing will be more closely interlinked in order to achieve a high level of quality at an early stage of development. The latest mathematical methods from multi-objective optimisation, which is one of the core competencies of the Institute of Industrial Mathematics, are used for this purpose. In this way, competing objectives such as energy efficiency, comfort and costs can be taken into account simultaneously and the safety of the system can also be guaranteed. The plan is to integrate the new processes into the dSPACE tool chain and evaluate them using an example from e.GO vehicle development.

"As consortium leader of the project, our goal is to take the next step towards a one-stop development environment for autonomous vehicles," explained Dr Rainer Rasche, Group Manager Test Automation at dSPACE. "The resulting tool chain enables the developer to design the parameters of an ECU for different, typical traffic situations and test them simultaneously in the simulated environments. This enables our customers to accelerate their development."

Dr Michael Riesener, Vice President Corporate Research at e.GO Mobile AG, said: "The simultaneous development and testing of new systems for our electric vehicles made possible by SET CPS continues to enable us to realise fast development times and design vehicles that are even more requirements-oriented. For this reason, we look forward to driving the research project forward together with our partners."

About e.GO Mobile AG

e.GO Mobile AG was founded in 2015 by Prof Dr Günther Schuh as a manufacturer of electric vehicles. The more than 450 employees on the RWTH Aachen Campus utilise the unique network of the campus with its research institutions and around 360 technology companies. Agile teams are working on various cost-effective and customer-orientated electric vehicles for short-distance transport. Series production of the e.GO Life started in March 2019 at the new e.GO plant in Aachen Rothe Erde.

www.e-go-mobile.com

About IFIM

The Institute of Industrial Mathematics was founded at Paderborn University to facilitate a direct link to the transfer of applied mathematics to industry. Together with partners from industry, in particular small and medium-sized enterprises, mathematical problems are identified and efficient solution methods based on the latest scientific findings are developed. By bringing science and industry together, significant progress can be made in scientific, economic and technological terms.

About dSPACE

dSPACE develops and distributes integrated hardware and software tools for the development and testing of ECUs. As a full-service provider, dSPACE is a sought-after partner and solution provider in many current development areas of the automotive industry, from electromobility and automotive networking to autonomous driving. The customer base therefore includes almost all well-known automotive manufacturers and suppliers. In addition, dSPACE systems are also successfully used in aerospace and other industrial sectors. With more than 1,700 employees worldwide, dSPACE is represented at its headquarters in Paderborn, with three project centres in Germany and by subsidiaries in the USA, Great Britain, France, Japan, China and Croatia.

Information on the funding programme can be found at: https: //www.leitmarktagentur.nrw/leitmarktwettbewerbe/

Informationsbasierte Optimierung von Operationsplänen / Information-Based Optimisation fo Surgery Schedules
 

In Germany, the healthcare sector is one of the most important branches of the economy and is subject to constantly rising expenditure. Hospitals and operating theatres in particular account for a large proportion of these costs. In order to provide better care for patients and reduce operating costs and overtime, a more efficient management of operating theatres is needed.

Project description
 

The Information-Based Optimisation of Sugery Schedules (IBOSS) project is researching the development of new efficient methods to improve work and patient flow in hospitals. We are working closely with our project partner Charité Berlin to develop these concepts and algorithms. One part of the project is the forward-looking analysis of the sub-processes involved in a hospital so that these can be modelled precisely. On this basis, we are developing algorithms for the calculation of operation plans in which optimisations are carried out at both a micro and a macro level. Particular attention is paid to the algorithmic treatment of stochastic influences such as delays within an operation or sudden emergencies. The solution approaches are based on the following techniques:

  • Optimal learning of classifiers in data analysis
  • Stochastic/robust resource-based project planning
  • Multi-objective optimisation and optimal control of Markov processes

The aim is to develop an adaptive, self-learning optimisation system that automatically detects deviations and developments within the changing operational environment. Finally, a first prototype of the system is to be tested and validated in practical application.

Sub-project "Multi-objective optimisation of dynamic models for operating theatres"
 

In addition to the sequence of operations, there are many other factors that influence the quality of operating theatre schedules. These include, for example, the distribution of staff and medication as well as the start times of the individual surgical steps. The corresponding decisions have an influence on several, usually conflicting, target functions. These include the quality of medical treatment, the ability to react to unexpected events, staff and patient satisfaction and economic factors. The set of optimal compromises between these criteria, the so-called Pareto set, must therefore be calculated.

The aim of the Paderborn sub-project is therefore to develop a dynamic model of the surgical process, which is then used in an optimisation algorithm that works in parallel with the real process. Competing objectives and uncertainties must be taken into account. Depending on the news situation, an operation planner can then select an optimal compromise from the Pareto set. Furthermore, the results can be used to improve the overall operation planning.

IBOSS is a co-operation with the following institutions: Zuse Institute Berlin (ZIB), FU Berlin, Charité

Project homepage: Information-Based Optimisation of Surgery Schedules

AstroNet, a Marie Curie Research Training Network funded by the European Union from 2008 to 2010, brought together mathematicians, engineers and astronomers from universities, government organisations and industry to jointly develop innovative new developments in the field of astrodynamics. The main focus was on optimising trajectories and orbit control in order to minimise fuel consumption and increase the range of possible missions.

Every year, shortly before the end of the Bundesliga, the teams and their fans tremble over the championship, participation in European competitions or promotion and relegation. The same question arises every time: Which positions in the table are still possible at the end of the season?

In order to answer this question with certainty, a tool has been developed which, taking into account the possible outcomes of all remaining matches in the entire league, calculates for each team which place is still possible at best, worst or under its own steam up to a given matchday and, in particular, at the end of the season.

Project leader (IFIM): Prof Dr Michael Dellnitz


Project partner:

Deutsche Post Adress GmbH & Co KG, Gütersloh
 

Deutsche Post Adress offers companies automated solutions for address updating and address research. On the basis of extensive data on past address determinations, we identify previously undiscovered structures in this data using abstract methods of graph analysis, among other things. The aim is to optimise the updating and research processes and provide the customer with more reliable results, faster throughput times and more cost-effective processes.

Project manager (IFIM): Prof Dr Michael Dellnitz
 

Project partners: Herbert Kannegießer GmbH, Heinz Nixdorf Institute, Fraunhofer IEM, CITEC (Paderborn University), Paderborn University



Due to the market and competitive situation, industrial laundries will have to work quickly and cost-effectively in the future. Sustainable savings in resources such as energy, detergent, water etc. are essential. At present, machines in a industrial laundry are adjusted individually and independently of each other based on user experience. A systematic, mathematically based analysis of the optimum machine settings and a holistic view at the level of the entire laundry have not yet been carried out. This results in great potential for optimisation. In addition, the handling of laundry requires a high level of labour and hygiene requirements are playing an increasingly important role. This applies both to the handling of soiled laundry by the laundry staff and to the hygiene verification of the clean laundry delivered, e.g. to hospitals.

The aim of the research project is to significantly improve the resource efficiency of industrial laundries. Self-optimisation methods and processes are to be used to significantly reduce energy requirements in particular. The ecological and economic benefits are to be significantly increased as a result.

Project leader (IFIM): Prof Dr Michael Dellnitz

 

Co-operation partners: CITEC (Bielefeld University), CoR-Lab, Heinz Nixdorf Institute, Paderborn University

The demands placed on the reliability, user-friendliness and resource efficiency of products and production systems are increasing in line with customer requirements in terms of quality and operation. In order to avoid high costs, energy consumption must also be reduced. There is great potential for optimisation in self-optimisation (SO) processes that integrate intelligent behaviour into the systems so that devices and machines can adapt independently to changing operating conditions. For example, self-optimising energy management in electric vehicles can distribute the available energy depending on the operating situation and taking into account competing objectives, such as maximising comfort versus maximising range. In this way, the available energy reserves are utilised efficiently and an optimum overall result is achieved.

The aim of the research project is to develop a set of tools that makes methods and procedures for self-optimisation available in a user-friendly way. Companies can thus be supported in integrating self-optimisation into the mechanical engineering systems of tomorrow.

In sub-project 2, the Institute of Industrial Mathematics is developing problem-adapted algorithms for solving multi-objective optimisation and optimisation control problems together with partners from the various innovation projects of the Leading-Edge Cluster. Furthermore, challenges and obstacles that arise in the application of mathematical methods in the innovation projects are systematically collected and analysed with a view to making them available in the long term. On this basis, (partially) automatable solutions are developed for frequently occurring applications, which are made available in an integration and solution pattern library. Finally, a modular system of training measures is also being developed, on the basis of which individual knowledge transfers to the cluster companies can be organised.

Project manager (IFIM): Prof. Dr.-Ing. Joachim Böcker

Project partners: AEG Power Solutions GmbH, Paderborn University

Project manager (IFIM): Prof. Dr.-Ing. Joachim Böcker

Project partners: HELLA GmbH & Co. KGaA, Behr-Hella Thermocontrol GmbH, Bielefeld University, Paderborn University