International Transportation (71) 1 | 2019 48 SCIENCE & RESEARCH Academics Projects in a nutshell Overview of selected mobility research projects Bringing autonomous driving to life I n a newly founded innovation network, researchers from the Fraunhofer Institute for Industrial Engineering IAO have joined forces with McKinsey & Company and a range of industry project partners to collaborate on the mobility solutions of tomorrow - from conceptual idea to finished prototype. The researchers of the Mobility, Experience and Technology Lab, or MXT for short, look at the various ideas that are emerging, and determine as quickly as possible which technologies, services and business models could catch on and which of them are unlikely to succeed in the marketplace. One solution, for instance, envisages voice-assisted services that draw on artificial intelligence, with car windshields turning into multifunctional displays. At the heart of this initiative to pursue the most promising ideas is Fraunhofer IAO’s Mobility Innovation Lab in Stuttgart. This modern research facility for prototyping and creative workshops already provides an insight into the mobility of tomorrow, featuring, for instance, a converted vehicle that interacts with pedestrians; an electric three-wheel scooter that hints at the future of sustainable inner-city mobility; and a futuristic car cockpit complete with modular dashboard, windows made of switchable glass, reclining seats, foldout tables and a pull-out monitor. One of the key roles of the MXT Lab is to carry out user studies, providing a first indicator of the viability of potential innovation opportunities. In one of the first such studies, the partners investigated whether the time freed up by autonomous driving might be suited to language learning. This information allows the researchers to draw conclusions that can then be fed into the automation of the driving experience in the future and the way that these vehicles are designed. www.iao.fraunhofer.de A converted car interacts with a pedestrian. Photo: Fraunhofer IAO New breakthroughs in research on super-batteries R esearchers at Graz University of Technology (TU Graz) in Austria have discovered a means of suppressing singlet oxygen formation in lithium-oxygen batteries in order to extend their useful lives. Stefan Freunberger of the Institute for Chemistry and Technology of Materials at TU Graz has been working on development of a new generation of batteries with enhanced performance and longer useful lives, and which are also cheaper to produce than current models. He believes that lithium-oxygen batteries have significant potential. In 2017 he uncovered parallels between cell ageing in living organisms and in batteries. In both cases, highly reactive singlet oxygen is responsible for the ageing process. This form of oxygen is produced when lithium-oxygen batteries are charged or discharged. The researcher has now found ways to minimise the negative effects of singlet oxygen, and his findings have been published. In his paper in Nature Communications, Freunberger describes the effect of singlet oxygen on what are called redox mediators, which can be reversibly reduced or oxidised. The work was carried out in collaboration with researchers from South Korea and the USA. Redox mediators play a vital role in the flow of electrons between the exterior circuit and the charge storage material in oxygen batteries, and also have a considerable impact on their performance. The principle behind mediators is borrowed from nature, where they are responsible for a host of different functions in living cells, including transmitting nerve impulses and producing energy. The researchers used density functional theory calculations to demonstrate why certain classes of mediators are more resistant to singlet oxygen than others. They also identified its most likely avenues of attack. These insights are driving forward the development of new, more stable redox mediators. “The more stable the mediators, the more efficient, reversible and long-lasting the batteries become,” Freunberger explaines. Besides deactivating redox mediators, singlet oxygen also triggers parasitic reactions, which compromise battery life and rechargeability. Freunberger identified a suitable quencher that transforms the singlet oxygen produced into harmless triplet oxygen, which occurs in air - an enzyme called superoxide dismutase blocks the formation of singlet oxygen in living cells. www.tugraz.at International Transportation (71) 1 | 2019 49 Academics SCIENCE & RESEARCH Training data for autonomous driving A utonomous cars must perceive their environment true to reality. The corresponding algorithms are trained using a large number of image and video recordings. For the algorithm to recognize single image elements, such as a tree, a pedestrian or a road sign, these are labeled. Labeling is improved and accelerated by understand.ai, a startup established by computer scientist Philip Kessler, who studied at Karlsruhe Institute of Technology (KIT), and his cofounder Marc Mengler. An algorithm learns by examples and the more examples exist, the better it learns. That is why automotive industry needs a large amount of video and image material in machine learning for autonomous driving. So far, objects on the images have been labeled manually by human staff. “Big companies, such as Tesla, employ thousands of workers in Nigeria or India for this purpose. The process is troublesome and time-consuming,” Kessler explains. “We at understand.ai use artificial intelligence to make labeling up to ten times quicker and more precise.” Although image processing is highly automated in large parts, final quality control is made by humans. Combination of technology and human care is particularly important for safety-critical activities, such as autonomous driving,” the founder of understand.ai says. The labelings, also called annotations, in the image and video files have to agree with the real environment with pixel accuracy. The better the quality of the processed image data, the better is the algorithm that uses these data for training. More information: www.kcist.kit.edu | www.understand.ai Using processed images, algorithms learn to recognize the real environment for autonomous driving. Graphics: understand.ai Particulate matter from aircraft engines affects airways R esearchers under the leadership of the University of Bern have investigated the effect of exhaust particles from aircraft turbine engines on human lung cells. The cells reacted most strongly to particles emitted during ground idling. It was also shown that the cytotoxic effect is only to some extent comparable to that of particles from gasoline and diesel engines. According to the World Health Organization (WHO), seven million people worldwide die as a consequence of air pollution every year. For around 20 years, studies have shown that airborne particulate matter negatively affects human health. However, the toxicity of the solid particles from aircraft turbine engines is still widely unresearched. Now a multidisciplinary team, led by lung researcher Marianne Geiser of the Institute of Anatomy at the University of Bern, together with colleagues from Empa Dübendorf and the University of Applied Sciences and Arts Northwestern Switzerland (FHNW), has shown that primary soot particles from kerosene combustion in aircraft turbine engines also cause direct damage to lung cells and can trigger an inflammatory reaction if the solid particles - as simulated in the experiment - are inhaled in the direct vicinity of the engine. The researchers demonstrated for the first time that the damaging effects also depend on the operating conditions of the turbine engine, the composition of the fuel, and the structure of the generated particles. The present study was published in the journal “Nature Communications Biology”. Particles emitted from aircraft turbine engines are generally ultrafine, i.e. smaller than 100 nm. By way of comparison, a human hair has a diameter of about 80,000-nm. When inhaled, these nanoparticles - like those from other combustion sources - efficiently deposit in the airways. In healthy people, the well-developed defense mechanisms in the lungs normally take care of rendering the deposited particles ineffective and removing them from the lungs as quickly as possible. However, if the inhaled particles manage to overcome these defense mechanisms, due to their structure or physico-chemical properties, there is a danger for irreparable damage to the lung tissue. The particles turned out to cause different degrees of damage depending on the turbine thrust level and type of fuel: the highest values were recorded for conventional fuel at ground idling, and for biofuel in climb mode. Bibliographic information: Jonsdottir HR, Delaval M, Leni Z, Keller A, Brem BT, Siegerist F, Schönenberger D, Durdina L, Elser M, Burtscher H, Liati A, Geiser M.: Non-volatile particle emissions from aircraft turbine engines at ground-idle induce oxidative stress in bronchial cells. Nature Communications Biology. 2: 90 (2019), DOI: 10.1038/ s42003-019-0332-7 Close-up of the turbine engine at the testing facility with the aerosol sampling probe in place Photo: University of Bern / SR Technics Switzerland AG International Transportation (71) 1 | 2019 50 SCIENCE & RESEARCH Academics Metallic 3D printing on track for automotive series production E nd of March 2019, the joint project “Industrialization and Digitization of Additive Manufacturing (AM) for Automobile Series Processes - IDAM” held its kickoff meeting in Munich, which was intended to pave the way for Additive Manufacturing to enter automotive series production. Specifically, the project partners - consisting of SMEs, large companies and research institutions - will transfer metallic 3D printing into an industrialized and highly automated series process in the automotive industry for the first time. In this project, twelve project partners are laying an important cornerstone to sustainably strengthen Germany’s technological pioneering role and the country itself as a manufacturing location. By integrating metallic 3D printing into the conventional production lines of the automotive industry, IDAM will enable them to replace cost and time consuming processes, such as the production of molds, and to meet the desire for product customization at no extra cost. Metallic 3D printing is being implemented at two locations: the BMW Group’s Additive Manufacturing Center in Munich and automotive supplier GKN Powder Metallurgy’s factory of in Bonn. There, the IDAM team is qualifying the AM technology for the specific requirements to produce identical parts as well as individual and spare parts on the basis of specific components. The targeted quantities speak for the signal character of the joint project: In the future, it should be possible to produce at least 50,000 components per year in mass production and over 10,000 individual and spare parts - at the highest quality and under extreme cost pressure - with the AM production lines. Individual modules can be adapted to the different production requirements thanks to the modular construction of the line and, if necessary, replaced. http: / / ilt.fraunhofer.de Structural optimized differential housing, jointly developed by GKN Powder Metallurgy and Porsche Engineering. Photo: GKN Powder Metallurgy Picture: Sander Weeteling/ Unsplash Traffic prediction system based on neural networks R esearchers of the Miguel Hernández University (UMH) of Elche have developed artificial intelligence solutions based on deep neural networks to predict traffic conditions using data from fixed sensors (such as loops) and connected vehicles. This new system makes it possible to predict traffic 15 minutes ahead of time. To carry out this study, researchers of the UWICORE laboratory, which belongs to the I3E Centre for Engineering Research of the UMH, have digitised and implemented on the SUMO traffic simulation platform, a real traffic setting corresponding to a 97-kilometre stretch from Spain’s A-7 motorway between the cities of Alicante and Murcia. They have worked with the collaboration of the Levante Traffic Management Centre, which provided data on all their traffic sensors from the chosen stretch over a 12-year period. This stretch has been chosen due to the high influx of traffic (Daily Average Intensity of 100,000 cars on some spots) and due to the high number of traffic sensors on the stretch (99 in total), which make it possible to accurately measure traffic with a frequency of one minute. With a selection of this data, researchers have developed a digital simulation setting which makes it possible to very accurately generate the traffic endured by the A-7 stretch for 10 days. With the digital traffic platform created at the UMH, researchers have developed techniques based on deep neural networks to predict traffic conditions 15 minutes in the future, using data from connected vehicles. Researchers have analysed how the insertion of the connected vehicle affects the accuracy of the traffic intensity, density and speed predictions. Their investigations have allowed them to prove that traffic prediction levels can be improved with data from just 4% of the vehicles, compared to when the prediction is done with data from the traffic sensors that are currently deployed in the relevant A-7 stretch. They also have shown that the merger of the data provided by the current traffic sensors with data from connected vehicles allows for an improvement of traffic prediction accuracy. According to Strategy Analytics data, in 2019, more than half of all manufactured vehicles worldwide will be connected vehicles. With their data, it is possible to learn the state of traffic and even predict it, without having to deploy and maintain traffic sensors as is done today. However, access to this data will have a cost, which means it is important for administrations and managers to know how many pieces of data they need to conduct their functions. The research of the UMH not only offers tools based on artificial intelligence for the characterisation and prediction of traffic conditions, but also make it possible to quantify the data necessary to be able to accurately predict traffic conditions. For example, the percentage of vehicles from which data is needed. www.umh.es