Etikettarkiv: Ericsson

Okategoriserade

Guldkorn från svensk forskning 2020

Trust in What? Exploring the Interdependency between an Automated Vehicle’s Driving Style and Traffic Situations. As the progression from partial to fully autonomous vehicles (AVs) accelerates, the driver’s role will eventually change from that of active operator to that of passenger. It is argued that this change will lead to improved traffic safety, as well as increased comfort. However, to be able to reap the benefits, drivers must first trust the AV. Research into automation has shown that trust is an important prerequisite to using automation systems, since it plays an important role in creating user acceptance and in generating a positive user experience. Moreover, for the purposes of safe AV operation, it is important that the user’s trust in the automation is appropriate to the actual capabilities of the system. One important aspect that can build user trust is to conveyvehicle capability, something which is commonly communicated via displays located in the cockpit of the vehicle. However, it has also been shown that parameters such as lateral steering also provide the driver with an understanding of the vehicle’s capability. Therefore, driving styles, or how the act of driving an AV should be conducted, may affect a user’s trust. However, little research has been conducted on the impact of driving styles in AVs in everyday traffic situations; that is, situations often encountered in a day-to-day driving context, such as stopping for a pedestrian at a zebra crossing or overtaking a moving vehicle. An experimental study with 18 participants was conducted on a realistic test course using a Wizard of Oz approach. The experiment included seven everyday traffic situations that the participants’ experienced with two different driving styles, Defensive and Aggressive driving style. The results show that characteristics of everyday traffic situations have an effect on the users trust in automated vehicles (AVs). Primarily due to perceived risks (for oneself and others), task difficulties and how the AV conforms to the user’s expectation regarding how the AV should operate in everyday traffic situations. Furthermore, the results also show that there are are interdependencies between situational aspects and how the AV driving behaviour conducts actions. Thus, the AV driving behaviour needs to be designed to operate differently depending on the traffic situation, to enable the user to create an appropriate level of trust, in relation to the actual performance of the AV. Finally, trust results from the information provided by the AV’s behaviour, what it explicitly communicates via displays, and how these factors relate to the driving context. Thus, a systems approach is necessary, in which the interaction between user and automation is key, but without neglecting the equally important contextual aspects. This study was funded by Vinnova, Sweden’s Innovations Agency, under grant number 2014-01411. The study was able to use the facilities and expertise of the full-scale test environment AstaZero through the open research grant, application number A-0025. Here you can find full paper, and for more information contact Fredrick Ekman at Chalmers (fredrick.ekman@chalmers.se) or read his licentiate thesis titled Designing for Appropriate Trust in Automated Vehicles that was publicly presented earlier this year. 

The Day 1 C-ITS Application Green Light Optimal Speed Advisory. Leveraging the growing communication capabilities between vehicles, infrastructure and other road users, applications under the C-ITS umbrella are expected to improve road safety, traffic efficiency and comfort of driving by helping the driver take decisions and adapt to the traffic situation. The Day 1 set of C-ITS applications, as defined by the C-ROADS platform build on mature technologies and are expected to be deployable and provide benefits in the short term, but what scientific evidence is there on their effectiveness and what gaps in knowledge are there? For the C-ITS Day 1 application Green Light Optimal Speed Advisory (GLOSA), these questions were addressed by a systematic mapping study (to our knowledge, the first such study to be published), conducted as part of the Nordic Way 2 project (co-financed by Connecting Europe Facility, CEF project 2016-EU-TM-0051-S), presented at the European Transport Conference 2019 and published in Transportation Research Procedia in 2020. Among the findings where that while there are many published studies evaluating GLOSA, the absolute majority collect data in simulation, focused mainly on observable effects for the equipped vehicle where fuel consumption and travel time were the most prevalent effects examined. Further, there was great variation in the effects observed (for instance, fuel consumption varied from no evident reduction to approximately 70% reduction between studies) providing little consensus in concluding the effectiveness of the GLOSA application. A possible reason for the big effectiveness variation is a lack of well calibrated models used in the simulations scenarios, especially with regard to driver and fellow road user behaviour and precision of traffic light phase shift prognoses. For more information contact Niklas Mellegård at RISE (niklas.mellegard@ri.se).

Making autonomous drive skilled in extreme situations. During 2020 Sentient finalised the development and testing of the S+ Split-μ Control function, that makes autonomous drive safe in the critical situation of braking in an emergency on split friction roads. Compared to traditional ABS, the braking distance could be reduced by up to 37% while maintaining stability. The function is available also for use in manually driven cars to aid the driver perform like expert drivers would in a split-μ situation. Watch this demonstration from the Colmis test track outside of Arjeplog. More information about safety functions developed by Sentient is available at the company’s website.

Ljuddesign som ökar tillit och minskar åksjuka i självkörande bilar. Hur kan ljuddesign höja användarupplevelsen i automatiserade fordon? Denna fråga har Volvo Cars utforskat de senaste två åren tillsammans med RISE och Pole Position Production. Projektet Ljudinteraktion i Intelligenta Bilar har tagit fram helt nya typer av gränssnitt där passageraren får information om bilens kommande beteende, samt vad i trafikmiljön som bilen fokuserar på. Signalerna låter bland annat snarlikt bilens naturliga ljud vid acceleration och fartminskning, men spelas någon sekund innan bilen agerar. Projektets studier har visat att signalerna ökar passagerarnas tillit till bilen, samt minskar åksjuka för en majoritet av passagerarna. I projektets avslutade del implementeras en prototyp av ljudgränssnittet i en Volvobil, vilket gör det möjligt att uppleva ljuden i verklig trafikmiljö. Resultat från projektet kommer presenteras vid ett seminarium hos SAFER i slutet av januari. Hör av er till projektledaren Fredrik Hagman på Volvo Cars för mer info (fredrik.hagman@volvocars.com), eller besök projektets hemsida. Projektet finansieras av Fordonsstrategisk Forskning och Innovation (FFI).

DI-PPP public and private partnership platform for quick and effective implementation of digital transport infrastructure: This pre-study is jointly financed by Drive Sweden and Trafikverket to accelerate the implementation of digital infrastructure in Sweden. The project uses the Trafikverket roadmap on connected and automated road transport system extensively to explore the synergies and to support the service development. The project defines the digital transport infrastructure from a system of systems perspective with the identification of key areas, action points, and expected achievements for the year 2021 – 2025. The project calls for both top-down and bottom-up approaches to build infrastructure that on the one hand enables applications and services fulfilling the mobility needs, and on the other hand, is built on an existing infrastructure with incremental advancement. The project calls for the establishment of a public and private stakeholder partnership platform that is long-term, proactive and progressive, with strong engagement and balanced investments among stakeholders to accelerate the infrastructure implementation. The results have been presented at the Drive Sweden thematic area digital infrastructure, and for more details and reports, please contact Lei Chen at RISE (lei.chen@ri.se).

Project CeViSS. Cloud enhanced Vehicle – intelligent Sensor Sharing (CeViSS) is a joint Drive Sweden project that has run from January to December 2020. The project was financed in part by Vinnova / Drive Sweden with partnership including Carmenta, CEVT, Ericsson, Volvo Cars and Veoneer. The primary goal of the project was to extend the previously established AD Aware Traffic Control cloud with functions to study and demonstrate how the central cloud platform can be used to collect and enhance critical traffic information before safely sharing it between automotive actors. The project successfully demonstrated how data registered by a Veoneer vehicle’s sensors, was collected, analyzed and enhanced in real-time on the central cloud level and then shared with the two project OEM partners; CEVT and Volvo Cars. Their connected cars could then take appropriate action and more precisely mitigate the hazard on their road ahead. The project also showed how the Carmenta Central Traffic Cloud could send instructions to the Veoneer and CEVT cars such as a recommended speed inside geofences (to be used by the Adaptive Cruise Control (ACC)) and search requests to look for specific symbols or texts (e.g., license plate numbers). Tests were also done where the Central Traffic Cloud had direct control of on-board cameras to start sending video when the Veoneer’s test vehicle approached an accident scene. Images or live video from the scene have the potential to give 112 operators and first responders a better understanding of the situation and help dispatch the right resources as well as make a more detailed planning of the rescue operation before arrival. A series of workshops was arranged during the project with representatives from two rescue organisations to get their response on the value of the technology. Both KatastrofMedicinskt Centrum (KMC) and SOS Alarm confirmed that when planning a rescue operation as well as when organizing the work at the scene it is important to collect as much information as possible about the accident area. Images or live video transmitted from a recent accident under strict control have the potential to improve rescue operations. As the sharing of sensor data in such a way have possible privacy concerns, the legal aspects was also investigated. The results of the legal study is documented in a separate report, added as an appendix to this document. The main deliverables from the project were live proof-of-concept trials performed at several occasions with final tests successfully completed at AstaZero test track, October 19, 2020. A film documenting these tests and explaining the project results was produced and a presentation held at a webcasted Drive Sweden event on December 1, 2020 concluded the project. The project has based its work on the cloud-based platform that was created in the project ”AD Aware Traffic Control” and further extended in the project ”AD Aware Traffic Control Emergency vehicles” and the following ”AD Aware Traffic Control – Advanced Cooperative Driver Assistance” project. The project used technology in Drive Sweden Innovation Cloud and its results will be integrated in this innovation platform for future use. For more information contact Kristian Jaldemark at Carmenta (Kristian.Jaldemark@carmenta.com).

Digital Twins Are Not Monozygotic – Replicating ADAS Testing Across Simulators. Testing in simulators is an essential component in cost-efficient and effective ADAS development. Without countless hours on virtual test tracks, arguing that an ADAS is safe for use on public roads will be practically impossible. However, how can we interpret issues that are detected in a simulator? Would they generalize to the real-world environment? Would they even generalize to another simulator? In a joint study with the University of Luxembourg, RISE used search-based software testing to identify safety violations of a pedestrian detection system in TASS/Siemens PreScan and ESI Pro-SiVIC. However, when replicating the same scenario in the other simulator, the researchers found that the results often differed substantially. Consequently, the researchers recommend future V&V plans to include multiple simulators to support robust simulation-based testing. Make sure the ADAS works safely in other simulators before hitting the real-world roads! The paper pre-print is available here, for more information contact Markus Borg at RISE (markus.borg@ri.se).

Nordic initiative for transport of passengers and goods by drone (NDI): The Nordic countries are joining forces to drive the development of drone transports for both goods and passengers. The Nordic Drone Initiative (NDI) will pave the way for new sustainable business models. It can be about air-taxis, autonomous courier services or new tourist concepts. NDI is co-financed by Nordic Innovation through their Nordic Smart Mobility and Connectivity program, led by RISE and consists of 16 partners from four Nordic countries including RISE, Katla Aero, Flypulse, Kista Science City, Mainbase, LFV and Region Östergötland from Sweden; VTT, Bell Rock Advisors, Robots Expert, Business Tampere from Finland; NORCE, Nordic Edge, UAS Norway and Drone Nord from Norway; and Gate21 from Denmark. The project reference group includes Norwegian Avinor ANS and Finnish ANS. The project is welcoming partners and will collaborate with NEA – the Nordic Network for Electric Aviation to jointly plan for short- and long-haul transports with electric aircraft. For collaborations, please contact Tor Skoglund at RISE (tor.skoglund@ri.se).

Testing safety of intelligent connected vehicles in open and mixed road environment (ICV-Safe): This project is a bilateral joint effort to identify safety-critical scenarios and to develop risk assessment and mitigation methods for intelligent connected vehicles (ICVs) by taking advantage of the large-scale open connected test environment in Shanghai. The project will conduct iterative case design, data collection, simulation, and open road test. The results will lay a foundation for the safe introduction of ICVs to minimize safety risks. RISE is coordinating the Swedish part with partners including Chalmers University of Technology, Alkit Communications AB, WSP AB, and FellowBot AB. The Chinese part is coordinated by Tongji University with partners including Research Institute of Highway (RIOH) Ministry of Transport, Chang’an University, Guangzhou O.CN International Technology Co., Ltd, Shanghai SongHong Intelligent Automotive Technology Co., Ltd., and Beijing Tusen Weilai Technology Co., Ltd (TuSimple). Through the project, the partners are also working actively with Swedish actors in China outside the project consortium to explore synergies for further research collaborations and innovation. For more details, please contact Lei Chen at RISE (lei.chen@ri.se).

CTS – Heterogeneous project. This project aims to investigate effects of autonomous vehicle in a mixed traffic environment, i.e., the traffic where automated vehicles share roads with different types of manually-driven vehicles. Effects on traffic flow and safety are the main interests of the project. An example of upcoming activities in the project is a driving simulation study, which is planned during January-February 2021. The study aims to investigate whether there is a behavior adaptation among human drivers when they share roads with automated vehicles. This project is funded by VINNOVA, and it is within the scope of CTS (The China Sweden Research Centre for Traffic Safety), which is an on-going collaboration within SAFER’s research program. Partners on the Swedish consortium includes VTI, Chalmers, Volvo Cars, and Volvo Group; and partners on the Chinese consortium are RIOH, Beijing Jingwei HiRain, Tsinghua University, and Tongji University. Link: Heterogeneous Traffic Groups Cooperative Driving Behaviours Research under Mixed Traffic Condition | SAFER – Vehicle and Traffic Safety Centre at Chalmers (saferresearch.com).

Drivers’ ability to engage in a non-driving related task while in automated driving mode in real traffic. Engaging in non-driving related tasks (NDRTs) while driving can be considered distracting and safety detrimental. However, with the introduction of highly automated driving systems that relieve drivers from driving, more NDRTs will be feasible. In fact, many car manufacturers emphasize that one of the main advantages with automated cars is that it “frees up time” for other activities while on the move. This paper investigates how well drivers are able to engage in an NDRT while in automated driving mode (i.e., SAE Level 4) in real traffic, via a Wizard of Oz platform. The NDRT was designed to be visually and cognitively demanding and require manual interaction. The results show that the drivers’ attention to a great extent shifted from the road ahead towards the NDRT. Participants could perform the NDRT equally well as when in an office (e.g. correct answers, time to completion), showing that the performance did not deteriorate when in the automated vehicle. Yet, many participants indicated that they noted and reacted to environmental changes and sudden changes in vehicle motion. Participants were also surprised by their own ability to, with ease, disconnect from driving. The presented study extends previous research by identifying that drivers to a high extent are able to engage in an NDRT while in automated mode in real traffic. This is promising for future of automated cars ability to “free up time” and enable drivers to engage in non-driving related activities. The study was conducted by Volvo Cars and RISE in collaboration between two FFI funded projects: TIC – Trust to Intelligent Cars and HARMONISE – Safe interaction with different levels of automation. A pre-print of the paper is available here, and for more information contact Jonas Andersson at RISE (jonas.andersson@ri.se). 

Remote Driving Operation (REDO) project. Remote driving operation or teleoperated driving can support deployment, operation, and testing of automated vehicles. With advancement in wireless communication technology, this has recently becomes more feasible. In the REDO project, we are looking at different technical and non-technical aspects related to teleoperated driving, which include 1) interaction with remote operator; 2) feedback mode from vehicle to remote operator; 3) system architecture; and 4) laws and regulations. Demonstration is also planned towards the end of the project. This is a 3-year project funded by VINNOVA. The partners in the project are: VTI, CEVT, Einride, Ericsson, Ictech, KTH, NEVS, and Voysys. Link: REmote Driving Operation – REDO | Vinnova. For more information contact Maytheewat Aramrattana at VTI (maytheewat.aramrattana@vti.se).

Human factors in remote operation of heavy vehicles. Currently, most highly automated vehicles still require the presence of a human safety operator in the vehicle, and it is evident that automated driving without human “fallback” might be distant. On the other hand, having a human operator in the vehicle jeopardizes major anticipated benefits of automated driving – productivity. This is especially evident when it comes to heavy automated vehicles. To bridge this gap, stakeholders are exploring teleoperations technology, which enables highly automated vehicles to be remotely operated if necessary. But remote operation comes with its own challenges, both from technical and human behavior perspectives. In this SAFER co-financed prestudy, Scania and RISE have identified potential safety challenges and research gaps related to human behavior in the context of remote operation of heavy automated vehicles. A general view of the human factors related challenges within the remote operation topic can be summarized by highlighting phenomena such as physical and psychological distancing, screen delays, network latency delays, inefficient interface designs, and human operator’s cognitive limitations. These are not exclusive to one single operational level, or application type, and are often interrelated. A larger body of scientific work can be found related to human factors in remote operation in other domains (e.g., robotics, aerial drones, military). Some of the findings from these domains can have value for the automotive domain, however, generally design requirements are not directly transferable between domains as there are domain specific challenges. An overall conclusion from the prestudy is that human factors in remote operation of highly automated road vehicles have been somewhat neglected by industry and research community. By providing an overall conceptualization of remote operation and its complexity, a theoretical framework, a state of the art overview, and a list of gaps and challenges, the expectation is that this pre-study will stimulate more activities in the area. The recently started FFI-project HAVOC is example of such an activity. The pre study was co-financed by SAFER and conducted by Scania and RISE. Link to final report, for more information contact Azra Habibovic at RISE (azra.habibovic@ri.se).

Task Force – Hygiene procedures in test with research persons. Since the rapid outbreak and continued global spread of the Coronavirus Disease (COVID-19) in 2020, aspects of much of our day-to-day life in society has been impacted – our workplaces are no exception. Due to the novelty of COVID-19 to health officials in Sweden and around the world, standardized guidelines on how to safely proceed with business activities that require the sharing of physical spaces and/or equipment between individuals has yet to be established. In anticipation of this pandemic being an ongoing issue, a task force was assembled to help address this gap. The SAFER task force was comprised of transport industry professionals in Sweden that have a role in conducting research and testing that would currently be deemed to place individuals at risk of contracting the virus if one of the involved actors were to be an active carrier of the virus. Therefore, the goal of this task force was to help establish a set of general guidelines to consider when attempting to mitigate the risk of contagion while performing research or testing activities at our respective corporate facilities. Questions related to “How can experiments involving test persons in vehicles, driving simulators, virtual-reality studios, or similar test facilities continue?”, “What safety procedures should we consider to introduce in order to ensure proper hygiene for the individuals involved?”, “Is it required for drivers to wear a face mask?”, and “How do we implement physical distancing provisions pre- and post-experiment interviews?” were addressed. Partners in the Task Force were VTI (coordinator), Volvo Group Trucks Technology, Autoliv, Veoneer, RISE and Scania. The project was co-financed by SAFER. For more information contact Arne Nåbo at VTI (arne.nabo@vti.se). 

Finansiering

Svenska startups i rörelse

Trots nästintill världsomfattande lågkonjunktur verkar det inte helt saknas pengar till startuppföretag inom mobilitet.

Göteborgsbaserade Annotell har precis fått tillskott på nära 70 miljoner kronor från bland annat Göteborgsbaserade Stenas Sessan och Ernström & Co [1].

Ett år efter att Einride lyckades dra in motsvarande ca 200 miljoner kronor har de nu fått in ytterligare ca 90 miljoner kronor från sina tidigare investerare Norrsken VC, EQT Ventures, Nordic Ninja V och Ericsson Ventures [2].

Egen kommentar

Att pengarna till dessa startups är lokala och från investerare redan till del knutna till bolagen ligger i linje med de analyser som tidigare presenterats. Värre kan det vara för bolag som försöker få in sina första investerare.

Källor

[1] Wauters, R. Tech.eu. Swedish startup Annotell raises €5.8 million to help autonomous vehicles see the world as it is. 2020-09-29 Länk

[2] O’Brien, C. Venture Beat. Einride raises $10 million to bolster autonomous trucking growth during the pandemic. 2020-10-01 Länk

Bussar, Personbilar, Tester och demonstrationer

Europeiska kollektiv-tester igång

Efter att ha tappat visst momentum under pandemins första våg har nu nyheter om tester med självkörande fordon för kollektiva transporter börjat trilla in igen.

På Djurgården i Stockholm har Keolis, Urban ICT Arena, Telia, Ericsson Intel och T-engineering precis börjat rulla publika tester med fordon som kopplas mot kontolltorn med 5G [1, 2]. Syftet med testet är att i en nära framtid kunna erbjuda säkra transporter med förare utanför fordonet. Testerna är en del av det pågående jätteprojektet SHOW i vilket RISE samordnar de svenska testsiterna.

Ett annat exempel hittar vi i Brașov i Rumänien, där det meddelas om ny satsning på test av självkörande fordon [3].

Egen kommentar

Efterfrågan på kollektiva persontransporter gick ned i spåren av pandemin och trafikbolag har haft det tufft. Att bolagen fortfarande visar på kraft nog för att fortsätta utvecklingen mot självkörande transporter är glädjande och pekar på att just de aktörerna tror extra mycket på teknikens potential.

Källor

[1] Keolis. Sweden: Keolis launches a new 5G autonomous electric vehicle trial in Stockholm. 2020-09-24 Länk

[2] Green Car Congress. Keolis launches a new 5G autonomous electric vehicle trial in Stockholm. 2020-09-25 Länk

[3] Fodor, S. Romania Insider. Central Romania city to test driverless bus. 2020-09-24 Länk

Bussar, Distributionsfordon, Inspiration, Övriga fordon, Sjöfart

Guldkorn från svensk forskning

Dessa guldkorn är bidrag från våra läsare – stort tack för det, och för all fantastisk forskning och utveckling som ni gör. Keep up the good work!

iQ-Pilot & iQ-Mobility. These are two recently finished projects co-funded by the Strategic vehicle research and innovation programme (FFI). The focus of the projects was development of new technology to realize flexible, energy-efficient transport solutions in cities. Several proof-of-concept prototypes have been developed and demonstrated, including autonomous buses and a smart coordination system for bus fleets. The research results were presented in a webinar earlier this week. These results are the joint efforts of Scania, Ericsson, INIT, Veoneer, Royal Institute of Technology (KTH) and Örebro University. 

Human interaction with autonomous minibuses. Tom Ziemke’s research group at Linköping University, in collaboration with researchers at VTI, will during the autumn start a new research project on people’s interaction with autonomous minibuses on campus. The research will focus on method development and empirical studies of how pedestrians, bicyclists and car drivers interact with the buses. A two-year postdoc position is available via this link (application deadline: August 5). For more information contact Tom Ziemke (tom.ziemke@liu.se).

GLAD – Goods delivery under the Last mile with Autonomous Driving vehicles. Small autonomous electric delivery vehicles (ADV) are expected to transform transportation of goods under the first and last mile. The advantages are increased transportation and energy effectiveness, but it is also important that these vehicles are safe and accepted in society. The aim of the GLAD project is to develop an initial knowledge base on efficiency, safety and human experience of ADVs for the first and last mile delivery of goods in Sweden, and on how to create a balance between these three aspects from a socio-technical perspective. To achieve this, the project will utilize Zbee vehicles that will be adapted in terms of vehicle design and autonomous vehicle behaviour, human-machine interface, teleoperation and vehicle management. The overall goal is to develop knowledge that accelerate introduction of new efficient goods delivery in our society and contributes to meeting the goals of Agenda 2030. This will be assured also by connecting a licentiate candidate to the project. The project is co-funded by Trafikverket and involves RISE, Halmstad University, Aptiv, Combitech and Clean Motion. It started in June 2020 and will run for ca 2 years. For more information contact azra.habibovic@ri.se.

Tactical Decision-Making in Autonomous Driving by Reinforcement Learning with Uncertainty Estimation. Reinforcement learning (RL) can be used to create a tactical decision-making agent for autonomous driving. However, previous approaches only output decisions and do not provide information about the agent’s confidence in the recommended actions. This paper investigates how a Bayesian RL technique, based on an ensemble of neural networks with additional randomized prior functions (RPF), can be used to estimate the uncertainty of decisions in autonomous driving. A method for classifying whether or not an action should be considered safe is also introduced. The performance of the ensemble RPF method is evaluated by training an agent on a highway driving scenario. It is shown that the trained agent can estimate the uncertainty of its decisions and indicate an unacceptable level when the agent faces a situation that is far from the training distribution. Furthermore, within the training distribution, the ensemble RPF agent outperforms a standard Deep Q-Network agent. In this study, the estimated uncertainty is used to choose safe actions in unknown situations. However, the uncertainty information could also be used to identify situations that should be added to the training process. The paper will be presented at the Intelligent Vehicles Symposium (IV) in October 2020, and a preprint is available on arXiv. The code that was used is also available on GitHub For more information, contact Carl-Johan Hoel (carl-johan.hoel@volvo.com) at Volvo Autonomous solutions. This work was partially supported by the Wallenberg Artificial Intelligence, Autonomous Systems and Software Program (WASP), funded by Knut and Alice Wallenberg Foundation, and partially by Vinnova FFI.

Autonomous Mapping of Unknown Environments Using a UAV. As part of the research conducted within the project LASH-Fire (Eu-Horizon 2020, No.814975), RISE supervised the work of Chalmers students developing an automatic object search for indoor environments using a flying drone. At the core of this system a reinforcement learning (RL) algorithm was implemented for the drone to navigate, detect obstacles, recognize objects and explore the environment. This machine learning (ML) project marks a starting point for further development towards an autonomous identification and surveillance solution in a wide range of study cases where cargo ships, like the ones studied in LASH-Fire, are an ideal target application. A modularized approach was used targeting research areas such as obstacle avoidance, object detection & recognition, simultaneous localization and mapping, etc. The exploration module was specially challenging and will require further work but the project in general was successful in providing a methodology and tools when using flying drones for indoor environments. The Master’s thesis was conducted by Erik Persson and Filip Heikkilä, and is available via this link. For more information contact boris.duran@ri.se

Projektet ESPLANADE, som började 2017 och avslutades sista mars 2020, handlade om hur man visar att ett automatiserat fordon är säkert. Det finns flera problem som måste hanteras för att man ska kunna göra en komplett säkerhetsargumentation. Projektets resultat inkluderar därför nya metoder för säkerhetsargumentation för en ADS, några av dessa är: 

  • En process för säkerhetsanalys samt designprinciper för interaktionen när en människa överlämnar kontrollen över ett fordon till en ADS eller tvärtom. Processen innehåller existerande metoder som sekvensdiagram, orsak-konsekvensanalys och felträd, men applicerade på människa-maskininteraktion istället för enbart tekniska system
  •  Hur man definierar den operativa designdomänen (ODD) för en ADS utgående från önskade användningsfall, vilket innebär en definition av parametrar inom vilka en ADS-funktion är avsedd att fungera, samt strategier för att säkerställa att fordonet håller sig inom sin ODD.
  • En metod (kallad QRN) för riskanalys och framtagande av säkerhetsmål. Till skillnad från vanliga riskanalysmetoder bygger den inte på analys av specifika situationer utan på definition av acceptabel frekvens av incidenter med olika allvarlig konsekvens, och en mappning av incidenter till olika klasser av konsekvenser. Säkerhetsmålen uttrycks så att man säkert hamnar inom acceptabla frekvenser.
  • Ett ramverk för formell och systematisk hantering av säkerhetskrav med en kombination av åtgärder under utveckling och under drift, bland annat baserat på modeller av osäkerhet.
  • Användning av metoden funktionsanalys för att distribuera beslutsfattande på en ADS-arkitektur samt framtagande av säkerhetskrav.
  • Säkerhetskontrakt och komponentbaserad design för att underlätta kompletthetsbevisning i kravnedbrytning, möjliggöra kontinuerlig produktuppdatering, samt kunna uttrycka säkerhetskrav för sensorsystem som inkluderar kamera, radar mm.

En publik rapport och länkar till de flesta av projektets publikationer finns på projekthemsidan.

Prepare Ships Project. Running for 26 months, the H2020 project “Prepare Ships”, funded by the European Global Navigation Satellite System Agency (GSA), was successfully started in December 2019. The 5 consortium partners, coming from 3 European countries have developed a machine learning based future position prediction for ships in order to avoid ship collisions and close quarter situations as well as reducing environmental impact by more advanced decision making. In a RTK (Real Time Kinematic) software solution, it will both exploiting the distinguished features of Galileo signals as well as combining it with other positioning and sensor technologies. It will use the next generation maritime communication techniques VDES and the new suit of IALA Standards (S100) on sea charts. The innovation developed during the project can make more autonomy of navigation feasible by exchanging future positions and allow eased decision making on ships, suitable to become an international game changer for the future of autonomous shipping. The demonstration and testing will be done onboard three different vessels in the Gothenburg archipelago. The project is coordinated by RISE with partners from across Europe, including SAAB, Lantmäteriet, Telko and Anavs. For more information check out our homepage, join our linkedin group or contact Johannes Hüffmeier at RISE (johannes.huffmeier@ri.se).  

How do you ensure safety of autonomous shipping? Today’s risk assessment methods, application of methods and models used in shipping are usually based on humans being directly in charge of ships, VTS, port controls, etc. and may not be sufficient to reflect and evaluate the complexities and inherent risks of introducing further automation and digitalization in the shipping domain. The introduction of smart ships will create traffic situations between manned and unmanned ships where on one hand decisions and actions are based on algorithms and on the other hand by a human operator where a large part of the decision making. Increasing the level of automation implies that the goal-based standards for shipping need to be based on a risk assessment that reflects the expected roadmaps towards more smart ships and so far, research on autonomous transportation has focused on other parts than the effect of introducing and mixing different levels of automation and only very basic standards have been proposed by classification societies, where DNVs standards [DNV, 2018] have two pages in the appendix on basic set-ups for testing and validation. The main objective of the RFAF project financed by Trafikverket is to analyse how autonomous navigation can be proven to be safe. The aim of the project is to perform a simulator-based risk identification for autonomous shipping traffic. Increasing the level of automation implies that the goal-based standards for shipping need to be based on a risk assessment that reflects the expected roadmaps towards autonomy. Based on two use cases, the routes Fredrikshamn-Göteborg and crossing of the Ljusterö fairway, relevant risks are identified based on ship simulations performed by mariners describing especially nautical challenges for more autonomous shipping resulting in a common risk model. The project lasts from January 2020-December 2022. There are 3 project partners with RISE as coordinator. For more information visit the project website or contact Johannes Hüffmeier (johannes.huffmeier@ri.se).

The SWEA-financed (Energimyndigheten) Data-driven Optimised Energy Efficiency of Ships is a national project involving 7 ship owners, 3 companies from the supply chain and RISE, lasting for 16 months. The data analysis of energy consumption is often complex and there are different driving forces for decisions. However, increased data collection can be unprofitable if you do not have methods to analyze the complex systems. Developments within machine learning provides new opportunities to develop both technically and economically powerful tools energy efficiency. Even today, to some extent, economic driving is applied, for example. eco-driving, however, the effect is in many cases limited as decision-making is more complex than the operator / navigator can see. Also, not always available incentives and motivation of individuals to reduce energy use. However, data collection is increasing both quality review and analysis are not performed to the same extent. Using the results of the project’s data collection and analysis, recommendations can be given about which tools which can be developed in a next step, such as: a) nudging, decision support system or autopilot for ECO driving, b) route optimization based on the ship’s accelerations and motions, and c) decision support based on statistics or real-time analysis of data to identify optimal operation (parameters such as sea state, current, speed, load condition, etc.). The objectives of the project are to: a) Achieve reduced energy use on the project’s vessels by 10–35% both at quay and in sea operations, b) Demonstrate potential with machine learning of operational data, and c) Demonstrate the possibility that better operational data may form the basis for the development of generic energy efficiency tools for smaller vessels in commercial traffic. For any details on the project, reach out to Johannes Hüffmeier (johannes.huffmeier@ri.se).

Photonics Private Public Partnership Roadmaps for EU’s next Framework Program Horizon Europé. The area of photonics for automotive applications is a significant area which includes not only photonics sensors for the EU defined topic Mobility and Safety for automated Road Transport. Photonics also plays a role in the path towards the targets of Zero Emission Road Transport, Clean Energy Transition, and the Industrial Battery Value Chain. The work of defining the Strategic Research Agenda (SRA) in the specific area of Photonics with EU industrial partners, universities and research centers is performed through the EU technology platform ”Photonics21”, which is funded by the EU commission. The current roadmap for Photonics was published in the document: “Europe’s age of light! How photonics will power growth and innovation, Strategic Roadmap 2021–2027” The section on Automotive and Transport can be found in section 3.9. The coordinator of the whole Photonics 21 is done by VDI Technologiezentrum GmbH in Düsseldorf, Link. We believe this is important as there are a lot of EU research money at stake. The current recommendation by the European Parliament for the whole Horizon Europe budget 2021 -2017 is €120 Billion. The research funding will be divided among many topics where Climate, Energy, and Mobility is one of the clusters. There is a large Swedish interest in the cluster and cooperation with industry is one important factor in the program. Most, if not all, of the European automotive industry are usually involved in at least selected programs.

Now, based on feedback from the new European Commission, the board of Photonic21 have decided to reshape the roadmap and as a consequence automotive & transport will henceforth be combined with the topics of climate and energy. Besides merging the different topics in one document, this gives us an opportunity to revise the previous document into something that we believe should support our industry even better, considering that the current document was prepared in 2018 and the present situation the industry is facing. We want to ensure that the guiding document capture the specific needs of the automotive industry. The aim of the work is to define the research topics of the Strategic Research Agenda (SRA) which will define the upcoming calls in the Horizon Europe program. 

We now invite comments on the current chapter and roadmap (provided in the link above). Determined by EU commission schedules this work has to be completed on 4 September, why we need your input no later than 24 August 2020. We ask for specific text suggestions and specific roadmap suggestions (compare with p. 140 in the above mentioned Strategic Roadmap). Please forward your suggestions to Jan-Erik Källhammer at jan-erik.kallhammer@veoneer.com. He acted as chair of the group Automotive and Transport in the current roadmap and now act as co-chair of the new group Climate, Energy, and Mobility together with Dr. Heinz Seyringer of V-Research GmbH in Austria. 

Kommunikation

Mer om 5G

Förra veckan skrev vi att Ericsson och Telia lanserat 5G i Stockholm. I samband med detta missade vi dock att rapportera att Tele2 (Göteborg, Stockholm, Malmö) samt Tre (Malmö, Lund, Helsingborg, Västerås, Uppsala och Stockholm) också lanserat varsitt 5G-nätverk [1, 2]. Spännande tider!

Kopplat till 5G-uppkopplingen så kan ni också passa på att se några illustrationer på hur 5G skulle kunna nyttjas för förarstödssystem (ADAS) i framtiden. Dessa exempel inkluderar varning för vattenplaning (5G aquaplaning alert), varning för dolda fotgängare (Hidden pedestrian alert) och förstärkt information om trafikskyltar (Augmented road sign information). Dessa premiärvisades i slutet av 2019 och har tagits fram i samarbete mellan Ericsson, Audi, TIM, Pirelli, Qualcomm, Italdesign, Tobii och KTH. 

Källor

[1] Tele2. Tele2 har lanserat Sveriges första publika 5G-nät. 2020-05-24 Länk

[2] Tre. ​Tre lanserar 5G i Skåne, Uppsala, Västerås och fler delar av Stockholm i juni. 2020-05-24 Länk

Infrastruktur, Kartor, Kommunikation

Kommersiell 5G i Stockholm

Telia har med hjälp av Ericssons 5G Radio Access Network (RAN) lanserat kommersiell 5G uppkoppling i Stockholm Centrum [1].

Områdena Norrmalm, Östermalm och Vasastan i Stockholm beräknas bli uppkopplade härnäst, och därefter satsar Telia på att utöka 5G till hela Sverige efter årets spektrumauktion.

Egen kommentar

5G teknologi förväntas spela en signifikant roll i utvecklingen av självkörande fordon. Framförallt så är applikationsmöjligheterna många inom uppkoppling av fordon och överföring av så kallad big data.

En funktion som kan möjliggöras med 5G är automatisk insamling av data kring vägkvalitet. Sensorer i uppkopplade fordon kan förse oss i realtid med detaljer om skicket på våra vägar. Företaget i-Probe, som är ett joint-venture mellan Softbank och två japanska företag Pacific Consultants (PCKK) och Oriental Consultants Global (OCG), ska ägna sig åt just detta i USA [2].

Källor

[1] Ericsson. Telia Company launches Ericsson-powered commercial 5G in Sweden. 2020-05-25 Länk

[2] Sawers, P., Venture Beat. SoftBank joint venture uses connected car data to improve U.S. roads. 2020-05-26 Länk

Kommunikation

Kommunikation i USA

Hittills har 5,9 GHz-frekvensbandet i USA använts exklusivt av den WiFi-baserade kommunikationsteknologin DSRC. I december 2019 godkände den amerikanska kommissionen för kommunikation (Federal Communications Commission, FCC) ett förslag som skulle möjliggöra för fordonsapplikationer baserade på mobilkommunikation (C-V2X) att använda vissa radiovågor i frekvensbandet 5,9 GHz [1]. FCCs förslag skulle också frigöra ett visst spektrum för användning av andra applikationer.

Förslaget har mötts av både positiva och negativa reaktioner. Under den senaste veckan har Ford uppmanat FCC att detta kommer innebära stora säkerhetsrisker [2]. De åberopar sin senaste forskning som visat att signalstörningar från WiFi sannolikt skulle hindra bilar från att effektivt kommunicera säkerhetsinformation med varandra, och föreslår att hela 5,9 GHz frekvensbandet bör användas enbart för säkerhetsapplikationer baserade på C-V2X. 

Den amerikanska konsumentorganisationen Consumer Reports är inne på lite samma spår som Ford. De menar att FCCs förslag är långt ifrån underbyggt av data och uppmanar FCC att dra tillbaka sitt nuvarande förslag [3]. 

Egen kommentar

Mer om C-V2X och DSRC problematiken kan ni bl.a. läsa om i våra tidigare nyhetsbrev, här och här.  Ni kan också läsa en artikel med Ericsson i focus här.

Källor

[1] FCC. NOTICE OF PROPOSED RULEMAKING. 2019-12-12 Länk

[2] Shankland, S., CNet. Ford warns FCC car-to-car communications proposal is unsafe. 2020-03-09 Länk

[3] Plungis, J., Consumer Reports. Ford Calls on FCC to Rethink Its V2X Auto Safety Plan. 2020-03-10 Länk

Bussar, Mobilitetstjänster

Två nya svenska projekt

Självkörande fordon på landsbygd. Projektet bedrivs av Ramboll, RISE, Trafikverket och kommunerna Skellefteå, Eskilstuna, Gotland och Lund och ska utreda möjligheterna att komplettera kollektivtrafik på svensk landsbygd med självkörande fordon givet kommuners lokala omständigheter och tekniska möjligheter [1]. Detta är en genomförbarhetsstudie som finansieras av Drive Sweden och Trafikverket och är en uppföljare till förstudien om samma ämne som slutfördes under våren 2019. 

REmote Driving Operation (REDO). Projektet bedrivs av VTI, CEVT, NEVS, Einride, Ericsson, KTH, Voysys och Ictech och ska undersöka olika aspekter av fjärrstyrda vägfordon, både personbilar och lastbilar [2]. Projektet delfinansieras av Vinnova, har en budget på ca 20 mijoner kronor och kommer att pågå i tre. 

Källor

[1] Ramboll. Självkörande bussar i landsbygd ökar tillgängligheten. 2020-02-04 Länk

[2] Ictech. Ictech deltar i ett av de största svenska forskningsprojekten kring fjärrstyrning av vägfordon. 2020-01-31 Länk

Kommunikation, Kooperativa system, Personbilar

VCC utforskar 5G i Kina

Volvo Cars (VCC) kommer inom ett samarbete med telekomleverantören China Unicom utforska och testa olika fordonstillämpningar av 5G och framväxande vehicle-to-all-teknik (V2X) i Kina [1]. 

Förhoppningen är att kunna identifiera potentiella förbättringar inom områden som säkerhet, hållbarhet, bekvämlighet och autonom körning. Exempel på tillämpningar inkluderar information om trafiken vid vägarbeten, möjligheten för bilar att hitta lediga parkeringsplatser lättare med hjälp av trafikkameror och harmonisering av hastigheten för att skapa en så kallad ”grön våg” vid trafikljus. 

Kina rullar för närvarande ut 5G i större städer och det här samarbetet kommer att hjälpa VCC att bekanta sig med lokala krav samt skapa en stark närvaro i landet i ett tidigt skede. 

Egen kommentar

Kampen om 5G har börjat på allvar, och många frågar sig vem som leder spelet? Det är svårt att säga, men enligt Ericssons VD Börje Ekholm så finns det ingen annan före Ericsson [2]. 

Om ni är nyfikna på 5G-tillämpningar i städer så kan ni läsa här hur Las Vegas gör sig redo. 

Källor

[1] Volvo Cars Media. Volvo Cars and China Unicom collaborate on 5G communication technology development in China. 2020-01-17 Länk

[2] Browne, B., CNBC. Ericsson CEO says there’s ’no one ahead of us’ on 5G — not even Huawei. 2020-01-21 Länk

Artificiell intelligens, Konferenser

Telematics Valley: AI in Automotive – Reality Check

Årets Telematics Valley-konferens handlade om AI inom fordonsindustrin. Här några korta referat.

The Great Math Gap of AI
 Carl Lindberg, AI Innovation Sweden
“Matematik, statistik och datavetenskap = AI”
Observation: många hatar matte i skolan och många är dåliga i matematik. Men för att förstå AI måste du kunna matematik och statistik, vilket krävs för att utveckla AI-team och deras kompetens.

AI essential component for automotive
Shafiq Urréhman, CEVT
AI-utveckling är tidskrävande och kräver mycket arbete med data. Datapreparation och märkning tar 80% av tiden.
Nästa stora steg för AI är att använda kvantdatorer för att öka beräkningskapaciteten.

AI based occupant sensing is the key to unleash a new level of functions
Henrik Lind, Smart Eye
Förarövervakningssystem krävs i fordon från 2023 av EU.

What we know that we don’t know
Mats Nordlund, Zenuity
Nyckelfrågor:
• Vad kommer andra trafikanter att göra?
• Bevisa säkerheten
• Framtida lagar och förordningar
• Träning av neurala nätverk
• Minska kostnader för sensorer
Forskningsområden:
• Prediktion av fotgängares rörelser och interaktion med fordon med hjälp av maskininlärning.
• Positionering och ruttprognoser

Automation
Sasko Cuklev, AB Volvo
Det finns stora vinster med automatiserade fordon för godstrafik:
• Ta bort föraren
• Ökad utnyttjandegrad av fordonen
• Minskat underhåll och reducerad bränsleförbrukning
• Plus säkrare och mer förutsägbara fordon
”Använd automatisering där det gör en stor nytta”
Affärsmodellen håller på att förändras. Numera beställer kunderna inte bara bilar, de önskar transporttjänster från A till B.
Flera piloter med automatiserade fordon pågår, bl.a i.
• Avgränsade områden, kalkstengruva i Norge
• Publika områden, hub till hub, 2 projekt pågår.
Vera är en transportlösning för gruvdrift, hamnar och motorvägar.

Egen kommentar: En bra presentation som visar applikationer för automatisering som inte är långt borta.

Autonomous driving in the Nordics –  Geofenced or SAE L5?
Hari Sentamala, Sensible 4
I närtid kommer autonoma fordon kunna köras inom avgränsade, geofencade områden. En säker autonom transport behöver kunna hantera alla väder- och miljöscenarier, där är vi inte idag. Sensible 4 arbetar med att använda redundans med hjälp av olika sensorer och kombinera och analysera resultaten från alla sensorerna så en säkrare autonom funktion kan fås.
Fälttester med autonoma fordon under vinterförhållanden ovanför polcirkeln har positionsnoggrannhet bättre än 18 cm uppnåtts.

Egen kommentar: En bra presentation med mycket humor. ”När SAE-nivån når 5 ändrar vi företagsnamn till Sensible 5”. Från en teknisk synpunkt ser de sig själva som ett sensormjukvaruföretag som använder många sensorer och uppkoppling i sina lösningar.

The importance of Data Quality and Governance for DAIR (Data and AI Ready)
Sofia Serafimovska, SAM Management Consulting
Affärsaspekter och KPIer är viktiga för att hitta och välja den information som krävs.

Machine Learning vs Software development – Verification & validation challenges
Lars Tornberg, Volvo Cars
Hur göra maskininlärning/AI säkert i verklig drift? Träningen av algoritmen kan inte återspegla alla möjliga scenarier.
Ett koncept är att använda en ”safety cage”. Analysera resultaten och validera modellen om de är trovärdiga. Föremål som modellen inte har tränats för, kan i ”safety cage” analysen tala om att detta är något nytt och resultatet därmed inte är säkerställt.

AI on the dark side of the moon
Peter Nordin, Semcon
Är AI ett hot? Ja! Elon Musk, Bill Gates, Stephan Hawkins tror det. Var försiktig med hur AI används. I första steget används AI för bra saker som att upptäcka cancer. Men redan idag är falska nyheter en verklighet. Psykologi och etik måste ingå i AI-utvecklingen.

Egen kommentar: Förmodligen måste EU och regeringar definiera etiska regler och lagstiftning om AI-lösningar. Peter gör en sammanfattning av science fiction-filmer som kan vara verklighet med AI släppt utan etiska aspekter. Då slutar det vara roligt.

Developing environmental model with ML from the ground truth data and scaling it in the cloud
Ulrich Wurstbauer, Luxoft
AD-validering kräver simulering i en virtual reality-modell ” Varför?

  • Fälttester på väg, 1 000 mil körning och ger nästan ingen data som krrävs för validering.
  • Re-simuleringar med hjälp av sensordata, 100 000 000 mil körning och ger ca 1% av data som krävs för validering
  • Simulator i virtual reality, kan ge 99% av nödvändiga data.

Men simulatorer kräver:

  • Modeller, med fokus på detektion, identifiering och prediktion.
  • Skalbarhet
  • En öppen dataplattform för att samla in, sortera och lagra data.

””Retail vision, applying AD approach for enterprise applications”
Atif Kureishy, ​​Teradata
Data + AI = Bättre svar och beslut.
Förutsäga försäljning med hjälp av människors beteende i butiker och använda information för att ändra butiken eller personalen för att förbättra försäljningen.

Egen kommentar: En känd applikation för AI att förutsäga försäljning eller kundreaktioner av olika marknadsföringsåtgärder.

AI risk, AI safety, AI ethics
Olle Häggström, Chalmers
EU har publicerat AI-dokument med etiska riktlinjer för tillförlitlig AI. Det är det första steget men behöver mer arbete för att vara användbar. Olle ser ett behov av regler för AI-applikationer.

Olles svar på publikens frågor:
Andra AI-områden kan vara att automatisera textilindustrin och föra produktionen närmare slutkunderna.
AI har svårt att ta över jobb med hög efterfrågan på social interaktion eller kreativitet.

”An inspiration map of AI in West Sweden.”
Erik Behm, BRG
AI växer inom transporter och fordonstillverkning. Life science, säkerhetsbranschen och finans växer också snabbt.

”Collaboration enabling driver-vehicle-infrastructure automation”
Edvard Brinck, Ericsson och Ola Boström, Veoneer
Uppgiften att att skapa förtroende för autonom mobilitet. Trafiksäkerhet är ett område som kan utvecklas med hjälp av AI och uppkopplade fordon.
2020 3 miljarder trafikanter, få automatiserade fordon >L2 och 1,4 miljoner dödsolyckor
2025 kommer L2+ att vara vanligt
2050 har samverkande säkerhet etablerats, det finns 6 miljarder trafikanter men dödsolyckorna har minskat till 0,7 miljoner.
Att förutsäga mobilitet för trafikanter är ett MobilityXlab-samarbete mellan Ericsson, Viscando och Veoneer.
Två megatrender är analys i realtid och tjänstefiering.

Egen kommentar: Ökad trafiksäkerhet är en bra användning av AI tillsammans med uppkopplade, samverkande fordon och trafiksystem.

How can AI and fashion help the exposed profession of truck drivers become safer?
Helena Iremo, Scania Group, Erik Tengedal, Imagimob
Uppgradering av säkerhetsvästen med uppkoppling och användning av ljus, intelligenta AI-algoritmer, accelerometer och gyro.

The journey to unleash the value of data with AI!
Robert Valton och Fredrik Moeschlin, AB Volvo
Beskriver arbetet hos Volvo med data och AI.

Develop AI cheaper and faster with collaborations
Hans Salomonsson, EmbeDL
Användning av syntetiskt genererade data för träning av AI.

Peter Kurzwelly, AI-innovation of Sweden.
Det finns en AI online-kurs på svenska. https://www.elementsofai.com/
Gör kursen! En uppföljning är på gång.

Sammanfattning

Det var många intressanta presentationer, och många om ”hur arbeta med AI i dina lösningar”. Min reflektion är att det har varit ännu mer intressant, om man tagit upp vad AI kan lösa och vad andra metoder kan lösa enklare för transportsystem och ta en bättre helhetssyn på ämnet.