On Wednesday 27th of February 2019, the Smart Bike project team gathered all the participants of Smart Bike one last time to officially draw the curtain on the 4 year project. Location of the event were the offices of BikeValley in Beringen. A final overview.
by DSP Valley
The Smart Bike project started in 2015 as cooperation between Flanders BikeValley and DSP Valley. Aim was to look for innovative opportunities in the overlap of electronics and the cycling domain. With a fresh action plan, clear deliverables and kindly funded by VLAIO, the team started its journey through this interesting world. Today, the team looks back on more than 35 identified innovation opportunities, resulting in 16 clear projects and 8 product launches.
Tested and evaluated
At the closing event the Smart Bike demonstrators were presented. A preliminary version of a bike tracker and a monitoring system to measure the position of the cyclist were the two projects in which the team had an active contribution.
During the project the tracker application was used by a professional cyclist whilst training in Lanzarote. The Smart Bike team members also used it on a daily basis to track their cycling progress. Positive about the bike tracker was that the Smart Bike team very early realized the need for this kind of app. Unfortunately the technology partner of the team turned out not to be the one needed. As a result other players on the market came up with a better performing design.
The system to measure the position of a professional cyclist in 3D turned out to be very useful in the wind tunnel testing facilities of BikeValley. But this system needs a very precise and accurate calibration every time it is used. The fact that the position sensors were stuck on a cycling shirt was also a problem. But solutions are possible since BikeValley is in contact with Bioracer, a speadwear manufacturer, and DSP Valley is involved in Smart-X, a European financing program for smart textiles.
Next to these two projects, eight other companies and organizations showcased their Smart Bike demonstrators. These range from a kit to electrify a regular bike over a sensor system to measure the quality of a specific bike lane to a system containing blockchain technology to encourage people to use their bikes more often. The event was setup in such a way that every participant passed by every demonstrator allowing ample time for questions, feedback and further contacts.
Does the official end of the project also mean the end of the cooperation between BikeValley and DSP Valley in this domain? Certainly not! The smart bike is definitely a major topic in the regular working of the BikeValley cluster. Companies with innovative ideas can always contact the former team members for advice, guidance and support. For DSP Valley, these activities are embedded in the Smart City program, where mobility as a whole is a large and important domain to cover. The links between both organizations, established during the project will continue to exist and will certainly result again in fruitful cooperation in the future.
If you are interested in the Smart Bike project or in the Smart Bike demonstrators, don’t hesitate to contact Bjorn Van de Vondel.
Solar Team Eindhoven is determined to not only build the most practical solar car in the world but also to drive the use of solar energy to the next level through the use of smart innovation. Verum sponsors the team’s software engineering activities with commercial Dezyne licences and technical support.
For over 30 years the phenomenal Bridgestone World Solar Challenge, a competition aimed at students the world over, has been at the forefront of driving innovation in solar-powered vehicles. Solar Team Eindhoven’s predecessors won the challenge in 2013, 2015 and 2017, a feat that the 2019 team plans to improve in the competition in October this year. A tough task that will take months of hard work and smart thinking. Verum will give the team an extra edge in the competition by sharing its knowledge of and experience in embedded software systems engineering with the team and by providing them with its unique, software toolset, Dezyne.
Next step towards a clean-energy future
Besides their ambition to outrun and outsmart the competition, the TU/e students want to contribute to solar energy development and the implementation of solar technology in our daily lives. To realise this ambition, they need to squeeze the maximum amount of energy out of the sun and convert it into motion by, amongst other things, designing highly effective and reliable embedded software in their car.
A first-time-right approach
Building software with Dezyne, a model-driven software engineering tool, is a first-time-right approach that results in a dramatic reduction of costs in the overall software lifecycle. Dezyne enables software engineers to specify, design, validate and formally verify software components for embedded, industrial & technical software systems.
Dezyne speeds up the development of control software by 300%, shortens development time by 20% and reduces field defects by 99.99%.
Based on the Eclipse IDE, Dezyne provides a complete environment for specifying, designing, exploring, verifying, generating and building model based software components. Engineers can easily understand, navigate through and communicate about their component models using a range of graphical representations, including state charts, event tables, system models and sequence charts.
Innovate in software
Dezyne integrates directly with conventional build environments and tools, enabling model verification, validation and code generation to be completely integrated with existing build processes. In brief, Dezyne speeds up the development of control software by 300%, shortens development time by 20% and reduces field defects by 99.99%. By using Dezyne to build their control software, the Solar Team Eindhoven will be able to rapidly innovate in software while also eliminating unpredictable and undefined behaviour, thereby achieving an unprecedented level of software robustness and reliability. This is one of the reasons the team is so eager to work with the Verum’s Dezyne tooling.
The Solar team Eindhoven aspires a future where daily life is powered by the sun. The Verum team is proud to support these ambitious students in reaching their inspiring and motivating goal.
Thursday, October 11, 8h45, a smooth drive to Eindhoven is the prelude to a rich and informative day at the Smart Systems Summit. To offer broad opportunities to matchmake, DSP Valley and Bits & Chips organize the event together. A report of a fruitful day.
by DSP Valley
The Van der Valk hotel, with its stately lounge with woodfire next to the entrance, is a marked change of the scenery for many DSP Valley attendants. Purpose of their day is an update on the latest evolutions in the smart systems market.
Arm sets the ball rolling
“To the day exact, 50 years ago, the first manned Apollo flight was launched. On board were 3 man and 1 flight computer with sensors and actuators. Quite literally an ‘edge’ computer with a mere 12 300 transistors, it needed sustained connectivity to earth to really be able to navigate the skies; the IBM computing system in Houston crunched sensor data, augmented with context, and uploaded indispensable additional guidance. A tiered, intelligent system – long before IoT. When 12 300 transistors can take you to the moon and back, think about today’s possibilities…”
After this short welcome by Dieter Therssen, CEO of DSP Valley, it is the Chairman of this Summit, Aad Vredenbregt, owner of ValOli and VP Bizdev at coMakeIT, who introduces the keynote speakers.
First on is Jürgen Jagst, senior manager automotive at arm. During his talk, Jürgen Jagst outlines the opportunities he sees for the smart industry in the society of tomorrow. Several macro-observations are discussed and documented, such as the escalating costs of designing solutions for volume markets, and the stalled cost-benefit of finer technology nodes since 28nm – cost per gate became essentially flat. There is a strong likelihood that after 10 years of customization wave (SoC and SiP for mobile market), we have embarked on 10 years of a ‘standardization’ trend with highly flexible devices for IoT applications. The silicon pendulum always seems to swing back.
We might have embarked on 10 years of a ‘standardization’ trend with highly flexible devices for IoT applications.
“We are living in interesting times”, Jürgen Jagst concludes ”, with innovation speed accelerating, China investing heavily in silicon technology, software more than ever becoming a key enabler, and – after the silicon industry consolidation wave – a consolidation ahead amongst platform and services players.” As if he was able to predict the very recent acquisition of the largest, independent Open Source company by IBM.
Microfsoft takes an open stance
After the 10 ‘o clock break Katrien De Graeve, IoT tech solutions architect at the Global Black Belt team of Microsoft brings the second keynote. She openly shares lessons learned from many actual implementations of IoT solutions. What are the challenges with digital transformation, what are the values realized when done properly? With Microsoft’s “Ignite” annual innovation showcase just 2 weeks earlier, a lot of new things and updates are introduced. Not in the least the investment commitment and strong emphasis Microsoft is putting on IoT solution enablers; cloud, security, AI, edge computing. Digital Twins are introduced as a concept to allow modeling the physical environment before connecting devices to that model. This facilitates factoring in ‘context’, ‘people’ and ‘spaces’ into the solution.
With Microsoft’s “Ignite” annual innovation showcase just 2 weeks earlier, a lot of new things and updates are introduced.
Katrien De Graeve also presents Azure IoT Central, a fully managed SaaS offering, which abstracts cloud intricacies from novel or novice users and helps them build compelling IoT scenarios. IoT Edge and Azure Sphere are introduced as platforms on the edge of the system, embracing – or at least supporting – Linux, and Docker containers. As perhaps best illustrated by its acquisition of GitHub, Microsoft seems to gradually take a more open stance and, on this summit too, Katrien De Graeve invites companies present to enter in discussion and join the ecosystem.
Indeed, Microsoft as well as arm are looking for new ideas and companies to work with, and after the keynotes and during lunch both Katrien De Graeve and Jürgen Jagst take their time to visit the exhibition and interact with the attendants.
After the keynotes are delivered, the participants split up and choose a presentation of their interest. The selection to choose from is quite diverse. There is Georgi Gayadadjiev from Maxeler. He kicks off in the Technologies for the IoT-track and demonstrates the advantages and the importance of cumulonimbus cloud systems and how that interface only captures the information you really need. After lunch, Robbert Lohmann from 2getthere explains how his company will integrate autonomous systems operating on public roads without safety driver or steward and uses his project at Brussels Airport as an example. In Smart Health, Nico Zeeders and Olesya Bliznyuk from Unitron tell the listeners what to pay attention to when bringing medical solutions to the European market. DSP Valley-members are also well represented in the pack of speakers. Peter Schepers from Itility takes IoT into the sky, Ramses Valvekens (easics) and Daan Gheysens (Robovision) tell us more on deep learning on FPGA in the case automated optical inspection (AOI) . And much more. The overall impression: a strong program with a high level of speeches. Or as one of the participants expresses it: “The quality of the speakers and their presentations is rock-solid”.
Members and start-ups take the floor
In between the different presentations, there is time for visiting the different booths and for networking. Next to members as Thaumatec, Itility and Achilles Design, there is also room for start-ups. Crodeon presents its hardware for IoT in the farming industry, Ivex shows how it programs the behavior of autonomous vehicles, Pozyx demonstrates its capabilities to track people and objects all over the world. And Epihunter exihibits the tool with which it monitors invisible epileptic seizures. Because the company has been in the news lately with this new technology, Ephihunter also gets the chance to give a presentation on the technology in the Smart Health-track.
Young blood showcases smart cars
To top off the conference, a display of the innovative power of young teams is presented in the form of 2 advanced vehicles from KU Leuven and the Eindhoven university of technology.
The “Stella” from TU/E is the prototype solar car that led to the ’Lightyear‘ vehicle presented by Arjo van der Ham in the mobility track. The team has been multiple winner of the Solare Challenge, and the car demonstrated raced 3 000 kilometers across the Australian outback to win the World Solar Challenge in October of 2017.
The “Pulse” from KU Leuven is this year’s contender in the Formula Electric for students. It is equipped with novel, artificial intelligence-based cooling technology and comprehensive telemetry, and won the 2018 First Prize for the design at the Formula Student Competition in the Czech Republic.
A great way to end the day, over a bite and a drink, having some concluding discussions, before heading home and let it all sink in.
The car industry is going through a fast evolution. According to some speakers at the Imaging & MEMS Summit, organized by SEMI in Grenoble (19 until 21 September), this evolution is not just a breakthrough, but rather a big bang! A big bang not happening in the future, but now! A summary of some interesting visions.
by DSP Valley
In line with this big bang statement, Luc Bourgeois, expert leader for ADAS and AD systems at Renault, explained how Renault today focuses on smart vehicles. The company more specificly focuses on autonomous driving vehicles, connected vehicles and their related mobility services, and on electric vehicles. The main challenges are in the autonomous driving vehicle. The basic technology is available, but the sensing is not at the right level yet for the performance and the robustness in an automotive environment. The quality of the sensing hardware shall be automotive quality, and not consumer electronics (CE) quality. Also obsolescence is an important aspect: automotive electronics should least for more than 8 years, whereas CE is typically less than 3 years. This is why car electronics is typically some technology nodes behind: nowadays 22-16 nm technology is being used in automotive, compared to CE which is already down to 7 nm.
Compared to human reliability, which is at a 10-8 level, a self-driving car is currently only at a reliability level of 10-3. Therefore, autonomous driving is not immediately expected to be available for everybody in all possible conditions. There is only a phased evolution towards autonomous driving: there will be a “scene expansion” from single-lane autonomous driving to multi-lane to city. On the other hand, this will be combined with a “delegation expansion” from safety benefits to full minds of experience. Renault has already incorporated these evolutions and expansions in the Symbioz concept car, where autonomous driving enables the car to become part of the daily living environment.
Compared to human reliability, which is at a 10-8 level, a self-driving car is currently only at a reliability level of 10-3.
Sense – Think – Act
The vision on the importance of the sensing system is shared by Lars Reger, CTO Automotive and Sr Vice-President at NXP. He emphasizes that a car is not just a brain on wheels, just like a human is more than a brain on shoes. For the computers in a self-driving car to perform correctly, they have to be fed with the right information, acquired by sensors: self-driving cars work according to the principle of Sense – Think – Act. The importance of the sensing system is proved by the fact that fatalities by smart cars are not caused by malfunction of the brains, but of the sensors!
Seeing around the corner
Further improvement of the sensing system is necessary, in order to come to a level of autonomy which can drastically reduce accidents. Therefore, NXP is developing a full sensor suite that goes beyond the traditional line-of-sight, to out-of-line-of-sight to line-of-feel. This can be realized with V2X (Vehicle-to-anything) sensors: based on the received information from infrastructure or from other users, these sensors allow seeing around the corner. Sensor systems for self-driving cars shall also be redundant: vision and radar are very complementary. Radar can be used in bad visual conditions. It can also be tuned from a wide to a very narrow and very accurate angle, by adapting its antenna system.
NXP is developing a sensor suite that goes beyond the line-of-sight, to out-of-line-of-sight to line-of-feel, so that your car will be able to see around the corner.
On top of that, LiDAR (Laser Imaging Detection And Ranging) will bring an important contribution to the sensing system, providing additional complementarity to classical vision cameras and radar sensors. At the summit, Filip Geuens, CEO of Xenomatix, presented an innovative multi-beam LiDAR solution, that proves to be more efficient than Flash LiDAR (which scatters in fog, or is confused by similar flashes) or Scanning LiDAR (which requires beam steering and high peak power). The presented multi-beam and chip-based solution offers a two-in-one 3D point cloud together with a 2D image, and is available as a windshield mounted device.
Vehicle to infrastructure
Klaus Habfast, Vice-Chair of the Grenoble Metropolitan Council, approached the deployment of the cars of the future from a different perspective. As a city authority representative, he highlighted that cities also play an important role in the introduction of self-driving cars. Sensors are at the core of smart cities. Placed in city roads or in city lighting they will deliver the necessary data to the self-driving cars through a Vehicle-to-Infrastructure link.
This said, Klaus Habfast believes that the future is not to self-driving vehicles, but to Connected Autonomous Vehicles (CAVs) that get a lot of their information from the smart city sensors. There is one consideration, he says. The infrastructure for CAVs cannot exist in isolation and should be shared with other applications; moreover, the cities have to ensure a citywide availability.
Are you interested to learn more about innovative technologies enabling the car of the future? Then the “International Conference on Automotive Industry in the Euregio” on November 29th in Alden-Biesen (near Hasselt) might be worth a visit. This evening conference brings inspirational contributions. Melexis talks about ICs for the electric powertrain, Punch Powertrain tells you more about silent electric motors for the electric car, FKA Aachen elaborates on automated driving: a safety gain or risk.
Interested? Then register today!
With more than 600 technology startups launching their business each year, Singapore is considered as one of the top startup hubs in the world. Additionally, the city-state is also ranking high on talent. No wonder it is home to the World Cities Summit. In the margins of this summit, DSP Valley went looking for a business intermediary for its members.
by DSP Valley
Singapore has become an established reference in the field of high tech and artificial intelligence (AI). The city-state is seen as the entry gate to South-East Asia and the world-class technology hub of the region. Singapore also counts 3 unicorn startup companies (a unicorn is a privately held startup valued at over $1 billion), Grab, Garena and Lazada Group. With a population of 5.6 Mio people, that is quite impressive. The European Union with almost 100 times more people counts less than 15,. Belgium has none, The Netherlands have/had one (Adyen – IPO June 2018).
With a population of only 5.6 Mio, Singapore counts 3 unicorn startup companies.
According to Forbes there are in total five key factors that contribute to Singapore’s pull, both as an innovation center and hub for regional operations. These five factors pertain to the framework that is offered by Singapore as a state. But there is more; the very essence of Singapore calls for creativity and innovation.
Housing and mobility
With a current population growth rate of 1 million per decade, Singapore needs new solutions for housing and mobility. For housing the city-state embraces new technologies to cope with the challenge of comfortably accommodating ever more people and maintaining the infrastructure over a long time. This results in Singapore being a front runner in Building Information Modeling (BIM), with 7D (sustainability) becoming the norm. BIM is the process that encompasses the entire development and management of the physical and functional information of a construction project. It are digital files describing every aspect of a building project and supporting decisions throughout a building life cycle.
Singapore is a front runner in Building Information Modeling, with 7D (sustainability) becoming the norm.
Concerning mobility, Singapore statistics are not bad: an average of 10 hours per person per year (hpppy) spent in congestion in comparison to 20 hpppy for the Low Countries [INRIX 2017 Traffic Scorecard]. Still, a lot of effort is invested in improving and preparing for the future. The comprehensive e-car sharing project rolled out with Blue-SG is a clear illustration of that planning for the future. And of how it brings business to innovative companies in Europe.
But Singapore does not stop there. Housing, mobility and integration are addressed at an even larger scale when we look at the plans for future towns in Singapore. Town centers will be transport hubs laid out in multiple levels. Upper underground levels will be used for commuter transportation to switch from bulk autonomous transport to the last mile transportation. Second and third levels under will be used for logistics operations, where during night time the provisioning of town shops and offices is catered for by autonomous vehicles. Here too are plenty of business opportunities for technology providers from around the globe to contribute to this vision. Reason enough for DSP Valley to take part in the trade mission to Singapore, organized by the Smart City Tech Alliance, and find a local partner.
One such outstanding local partner is IPI (Intellectual Property Intermediary). The organization was established in 2011 under Singapore’s Ministry of Trade and Industry. IPI’s goal is to enable enterprises to grow their businesses through technology and innovation. The organization does that by connecting Singapore-based businesses with local and overseas solution providers from both public and private sectors. Additionally, they facilitate partnership to bring new products & services to the market. IPI can do this because over the years, they developed a global network of technology partners and a technology marketplace featuring innovative technology from a wide range of industries. IPI also brought local startups and SMEs to our meeting, looking for opportunities to do business.
If you are looking to expand your business or if you have technology or solutions that fit in smart, eco-friendly buildings, or mobility solutions of the future, then Singapore might be a beach-head market for your innovation. An interesting tool you can use to get foot on shore in the city-state is the Gov-PACT Programme. By login in to this tool, you get informed on Singapore’s calls on innovation and you can submit your applications. If that is not exactly what you are looking for, let us know, so we discuss together the inroads DSP Valley can create for you.
The Gov-PACT Programme informes you on on Singapore’s calls on innovation.
Interested in first hand information on Singapore? Don’t hesitate to contact us
Tusk IC, the latest spin-off from KU Leuven develops chips for millimeter wave (mmWave) frequencies. These chips are essential for the new generation of radars for self-driving cars and for the new 5G standard and its ultra-fast data communication.
Tusk IC is founded by four electrical engineers, who obtained their PhD from KU Leuven in the ESAT-MICAS research group. The company receives investments in the form of technology IP and capital from KU Leuven and the Gemma Frisius Fund. The company is located in Antwerp.
The time is ripe to make mmWave silicon chips available for the semiconductor industry.
The miniaturization of silicon ICs (integrated circuits) ensures that very high frequencies can be processed in very compact, affordable chips. This makes a whole range of new applications possible. The most important markets for Tusk IC are the automotive industry (radars for self-driving cars), telecom (5G, wireless HDMI and VR) and industrial quality control. During their PhD, the founders specialized in the design of millimeter wave and terahertz chips in silicon technologies such as CMOS and SiGe.
According to CEO Wouter Steyaert, the interest and reactions are very positive: “With Tusk IC we fill a void in the current electronics landscape. There is a lot of interest in the development of mmWave chips for 5G and autonomous vehicles, but few engineers have concrete experience with it. Our complementary team has pioneered these frequencies during the research at KU Leuven. With Tusk IC we offer this knowledge and experience to both established multinationals and emerging companies in the mmWave market.”
Professor Patrick Reynaert (ESAT-MICAS, Department of Electrical Engineering) explains: “About 10 years ago, my team at KU Leuven started researching mmWave silicon chips. We developed the building blocks for 28 GHz to 600 GHz, used for 5G and radar. My research group has set up a reliable methodology, which offers an answer to the many challenges that come with these high frequencies. This has led to mmWave circuits whose measured results are closely related to the simulations. In my view, the time is now ripe to make this expertise available for the semiconductor industry. ”
With the advent of autonomous vehicles and 5G, the demand for mmWave chips will only increase. Tusk IC is determined to be the go-to company for mmWave IC design.
Our modern society suffers from a series of problems regarding mobility. Commuters regularly face seemingly endless traffic jams and queues to get from their homes to their workplace and vice versa. Policy makers often refer to “Smart Mobility” as a possible solution for these problems. But why are we not there yet?
by DSP Valley
A recently conducted European survey has shown that we have the technology for intelligent infrastructure, we know how to build Smart Cars, and we are able to produce Smart Bikes. So, the technology is there, but getting to Smart Mobility requires more. A workshop, organized by DSP Valley and hosted on the Advanced Engineering fair on May 17th, tried to come up with some answers. Three presenters looked at the mobility problem and suggested a way to tackle it, looking from their perspective. They all focused on the question ‘What is additionally needed to embrace Smart Mobility?’
A real demand for applicable solutions
First in the line-up was Tim Asperges, Mobility expert at the city of Leuven. In very clear words he elaborated on the mobility of the city. What are the different problem areas ? How were they addressed in the past and how should we tackle them now and in the future ? And most importantly: how can technology be of help? At some points, he explicitly solicited for technological inputs asking applicable solutions.
The city of Leuven asked for applicable solutions in Smart Mobility.
Detection of traffic movements
The second speaker, Jan Cools , CEO of Be-Mobile, took a more scientific approach of the problem. He elaborated on their developed techniques to measure and analyse the problem. By means of anonymous data, Be-Mobile detects movements in the traffic streams. They detect traffic jams, slowing traffic or other incidents and record these informations. Afterwards, conclusions are drawn and solutions sought by the use of statistical models. This way of working is indispensable for automatic driver support and even for autonomous driving.
Drones monitor critical traffic situations
The final presenter, Frank Vanwelkenhuyzen, founder of DroneMatrix, focused on a new way of monitoring dangerous or critical traffic situations. By using a drone to capture all kinds of data of the situation to analyse, they are able to derive different interesting metrics to analyse the traffic safety and other mobility aspects. The use of technology makes it possible to eliminate the necessity of human observers, hence reducing the possibility of errors. An ingenious software model at the heart of the system takes care of the data processing and visualization.
Using different ways to analyse, measure, evaluate the mobility situation continuously will be key.
Although we had three entirely different presenters and presentation topics, they all provided food for thought in the journey towards smart mobility. It will be key to understand and use different kinds of ways to analyse, measure, evaluate the mobility situation continuously, if we want to reach the end goal: smart mobility everywhere for everyone.
The coming of autonomous systems doesn’t just mean self-driving cars. Advances in artificial intelligence will soon mean that we, for example, have drones delivering medicines, crew-less ships navigating safely through busy sea lanes, and all kinds of robots assisting us.
As long as these autonomous systems stay out of sight, or out of reach, they are readily accepted by people. The rapid and powerful movements of assembly-line robots can be a little ominous, but while these machines are at a distance or inside protective cages we are at ease. However, in the near future we’ll be interacting with “cobots” – robots intended to assist humans in a shared workspace. For this to happen smoothly we need to ensure that the cobots will never accidently harm us. This question of safety when interacting with humans is paramount. No one worries about a factory full of autonomous machines that are assembling cars. But if these cars are self-driving, then the question of their safety is raised immediately. People lack trust in autonomous machines and are much less prepared to tolerate a mistake made by one. So even though the widespread introduction of autonomous vehicles would almost eliminate the more-than 20,000 deaths on European roads each year, it will not happen until we can provide the assurance that these systems will be safe and perform as intended. And this is true for just about every autonomous system that brings humans and automated machines into contact.
If deployed tomorrow, existing self-driving cars would have many fewer accidents than those driven by humans. But this doesn’t mean that people are ready to hand-over the steering wheel. We tolerate many thousands of deaths on the road every year, but the first crash involving two full-autonomous vehicles that results in a fatality will be/are now headline news all over the world. And then what? Will there be a public outcry? Will gangs come with pitchforks to smash the machines? Will self-driving cars be like the Hindenburg disaster and airships? Autonomous vehicles, indeed all autonomous systems, need to be made safe enough so that people trust them. The destination, therefore, is clear; the route, however, is a difficult one. The Safer Autonomous Systems ITN project is designed to get us to our destination, safely.
Not based on evolution
Until now, safety assurance has been integrated into the design processes, based on safety standards and demonstrating compliance during the system’s test phases. However, existing standards are developed primarily for human-in-the-loop systems, where a human can step in and take over at any time. They do not extend to autonomous systems, where behavior is based on pre-defined responses to a particular situation. What’s more, current assurance approaches generally assume that once the system is deployed, it will not learn or evolve. On the one hand, advances in machine learning mean that autonomous systems can be given the potential to learn from their mistakes, and the mistakes of all the systems they are connected to, making their abilities to operate safely infinitely better than previous generations. On the other, machine learning means more uncertainty about how the system will decide to react to a particular circumstance in the future, making safety assurance a hard task, which can only be accomplished by a highly-skilled, interdisciplinary workforce.
Machine learning means more uncertainty about how the system will react to a particular circumstance, making safety assurance a hard task.
Are you ready yet, to take a seat on an autonomously controlled airplane? If you hesitate to say “yes”, then you are tacitly acknowledging the need for a training and research program such as the Safer Autonomous Systems ITN.
Safe under all conditions
The main objective of the Safer Autonomous Systems (SAS) project is to identify ways that we can establish people’s trust in autonomous systems by making these systems demonstrably safer. In order to achieve this objective we have identified three challenges to be addressed by the early-stage researchers (ESRs) in their 15 individual research projects. This simply-stated objective, and the interdisciplinary needs required for its realization, is of such complexity that we see a large training network involving some of Europe’s flagship companies – such as Bosch, Airbus and Jaguar Land Rover – together with leading European universities – like KU Leuven and the University of York – as the best way to tackle these challenges, which are briefly described as follows:
Increased autonomy, by definition, means a significant reduction of the time during which a human is involved in the system’s decision making, thereby reducing the residual control afforded to humans. Studies have shown that it may take minutes for a non-actively involved human operator (e.g. a passenger in a self-driving car) to take over control in case of an emergency. Moreover, just putting a self-driving car to a full stop on a busy high-way by removing its power (so-called fail-stop behavior) is definitely not a safe action. In contrast, an autonomous system should be fail-operational (perhaps with reduced functionality) under all circumstances, monitor its own safety and make its own decision about a sensible and safe reaction. The challenge therefore is to design autonomous systems in such a way that they remain safe under all conditions, even in the case of component failures.
Virtual model-based testing
Testing is the most intuitive way to reveal unsafe behavior. However, autonomous systems must operate in a near infinite range of situations. When we test autonomous systems, we must therefore systematically determine which range and diversity of situations should be simulated and tested. We need to test them on roads, in the rain, and with people in the way. We need to test them when they’re in intermittent supervisor contact and when they’ve got an unbalanced wheel. And we need to test them in all the possible combinations of those cases. Testing autonomous systems in the field is clearly too costly and too time consuming and might even be harmful for the system or its environment. Hence, virtual model-based testing is the only viable option. However, breakthrough solutions are required to guarantee the rigor of our virtual testing and to optimize its overall coverage.
The main objective of the Safer Autonomous Systems (SAS) project is to identify ways that we can establish people’s trust in autonomous systems by making these systems demonstrably safer.
Three sub-objectives to achieve trust
More autonomy is possible only through new technologies, e.g., machine learning, for which no accepted safety-assurance strategies currently exist. Legacy experience as well as established standards and regulations are lacking. Implicitly or explicitly, current safety-assurance practices and safety standards assume that the behavior of the system is known at the design stage and can be assessed for its safety prior to system deployment. As autonomous systems might learn and evolve over time, this is no longer possible. This means that meeting the current safety standards for autonomous systems is either impossible to do or completely insufficient to assure safety throughout the life time of the system.
To achieve the main Scientific/Technical (S/T) objective of trust in autonomous systems by overcoming the 3 challenges we have decided on three sub-objectives that will be the aims of the project’s 3 research Work Packages (WPs):
Objective 1: To integrate guaranteed safe behavior directly into the architecture/design of the autonomous system.
Objective 2: To prove by model-based safety-analysis techniques that the behavior of an autonomous system remains safe under all possible conditions.
Objective 3: To ensure that the safety-assurance strategies that combine the architectural/design measures with the evidence allow us to have trust in the autonomous system, which is very likely to be learning and evolving.
For more information on the Safer Autonomous Systems ITN project, contact prof. dr. ir. Davy Pissoort or prof. dr. ing. Jeroen Boydens.
Imec has fabricated an innovative type of solid-state Li-ion battery. This battery is a milestone on the roadmap to surpass wet Li-ion battery performance and can become a contender to power tomorrow’s fast-charging, long-haul vehicles.
The future of mobility will be largely electrical, powered by fast-charging, safe, and compact batteries. Today’s rechargeable Li-ion batteries have some room for improvement, but not enough to allow vehicles sufficient range and autonomy. Imec’s researchers are working on a next generation of batteries, replacing the wet electrolyte with a solid, in order to increase the energy density of the cell.
Recently, imec developed a solid nanocomposite electrolyte with an exceptionally high conductivity of up to 10 mS/cm and with a potential to increase this even further. With this new electrolyte, imec has now made a prototype battery. The electrolyte was applied into the battery cell as a liquid precursor, and solidified afterwards. The prototype battery achieved a volumetric energy density of 200 Wh/liter at a charging speed of 0,5C (2 hours). A milestone on the roadmap to surpass wet Li-ion battery performance and reach 1000Wh/L at 2C by 2024.
“Our results show that we can make solid-state batteries that have the potential to reach the capabilities of wet batteries, and this using manufacturing processes similar to those for wet batteries,” says Philippe Vereecken, principal scientist and program manager at imec, “But unlike wet-batteries, our solid-state batteries will be compatible with metallic lithium anodes with a target of 1,000Wh/liter at a charging speed of 2C (half an hour). This, together with their longer lifetime and improved safety, makes them a promising compact battery technology for tomorrow’s long-range vehicles.”
Imec’s solid-state batteries are promising for tomorrow’s long-range vehicles.
To further improve the battery performance, imec is looking into combining nanoparticle electrodes with its solid nanocomposite electrolyte. Imec uses ultra-thin coatings as so-called buffer layers to control the interface between the active electrode and electrolyte. This technology can also be used to improve the performance of standard liquid cells and even for all-solid-state batteries with pressed and sintered inorganic electrolytes.
Bringing innovative battery technology to fruition and transfer it to the market will require the involvement and commitment of the world’s major material suppliers and battery producers. Therefore, imec performs its battery R&D as a collaborative program for open innovation to which it invites all interested parties.
Watch the video: