Enhancing Automated Driving through IoT

Bram van den Ende
Project Manager
TNO

Abstract
Automated driving requires the vehicle to establish a high degree of situational awareness as well as smartness in its decision making algorithms, while maintaining functional safety in all situations. A vehicle’s sensor suite is essential in this respect as it generates the vehicle’s world model, but has range limitations and also comes at a price. The connected vehicle concept allows to exploit the availability of information external to the vehicle to basically enhance the vehicle’s world model (virtual sensors) facilitating tactical level path planning. It also makes it possible to involve the vehicle in automated driving oriented service concepts. If we then take Internet of Things (IoT) concepts to the drawing board, the emerging IoT ecosystem becomes and enabler and quite powerful functionalities can be developed. In the European AUTOPILOT project TNO works together with partners in the Dutch Brainport Pilot to demonstrate and evaluate how automated driving can be accelerated and enhanced using connected car and IoT technologies. In our presentation we will explain how these concepts are turned into real applications, for example to conduct smart platooning on national roads. We will explain how the vehicles are wired into the Brainport IoT ecosystem set up in this project and how vehicles make use of a cloud based Platooning Service to get engaged in highway platooning. Sensor data from roadside cameras is processed to generate data with which the platoon’s world model can be expanded. We will show how this enables a formed platoon to better anticipate traffic situations further ahead. We will also discuss how AUTOPILOT is hooked into international standardization on IoT protocols and data models and present which application enhancements are on our project roadmap.

Biography
Ir A.H. (Bram) van den Ende has a degree in Electrical Engineering (1989) and has been employed at TNO almost the entire period (except from short periods at Philips (1990) and at Siemens (1999). Bram has lead numerous research based consultancy and innovation projects in the ICT domain since 2000. His main focus is on digital infrastructures, particularly in the context of national or international policy making or policy issues, but he also has had management responsibilities in FP7, H2020, Eurostars and EIT innovation actions. Specific topics are broadband access developments, spectrum issues, net neutrality and the application of mobile networking (5G) in Smart Mobility and PPDR verticals. Current engagements in the smart mobility domain are AUTOPILOT and 5G-MOBIX (new).

Internet of things and smart mobility

Martin Donoval
Director of R&D projects
POWERTEC ltd.

Abstract
Internet of Things (IoT) is the internetworking of physical devices, vehicles, buildings and other items embedded with electronics, software, sensors, actuators and network connectivity that enables these objects to collect and exchange data. IoT represents the next step towards the digitization of the society and economy, where objects can be interconnected through simply attachable and low power consumption communication networks and report about their surrounding environment. Mobility is a strong domain of industry with potential to employ IoT solutions in huge extent with many possible use cases. Development of IoT systems oriented towards smart mobility is therefore a big and challenging field. Network of IoT systems is based on low power consumption and long lifecycle of the products as the purposes of use can require years of operation in e.g. hermetically sealed casing. In specific cases, an intercommunication of IoT nodes is required. Design, development and production of such systems requires experience and specific skills in different areas, as are the power management, communication, self-test, antennas, casing and others. Among different IoT systems there are many subgroups of mobility sensing elements – static or dynamic, with individual connection to network or mesh connection, oriented towards monitoring of traffic, parking, position of carrier etc. The speech will be devoted to different attitudes to electrical design of static sensors and dynamic sensors based on example of comparison of static parking occupancy sensor for ground integration with at least over 5 years functionality and an hobby unnamed aerial vehicle (UAV) monitoring in 3D and intercommunication of the aerial vehicles.

Biography
Martin Donoval is experienced in area of electronic circuits and application level. Besides development of electronic devices for IoT applications and integration of devices into smart-systems he is responsible for R&D projects management in a field of microelectronics applications - bio-electronic smart-systems, mobility systems and other end user electronics. He participates in implementation and testing of new prototypes and dedicated electronics and prepares new testing routines. He finished his master and PhD theses at Slovak university of technology in Bratislava, Slovakia. He participates at the education process at STUBA.

A new ADAS Chip Design in 22 nm FDSOI Technology for Automotive Computer Vision Applications

Jens Benndorf
Managing Director, COO
DreamChip Technologies

Abstract
The presentation explores the European collaboration for a System-on-Chip development for Advanced Driver Assistance Systems (ADAS) in a brand new 22nm FDSOI low power technology implementation from Global Foundries. The presentation will elaborate on the history of the project, the partnership of the German consortium, the technology itself as well as on the target System-on-Chip implementation and the different applications within future ADAS devices. According to market analysis’s more than 60% of the cars sold today are equipped with different driver assistance systems and this number will grow over next years further. Because of growing demand, but also because of the very fast growing complexity and performance requirements of such systems cost reduction and power consumption become more and more important. Both requirements are addressed by the new architecture. A special focus will be put on HW/SW codesign trade offs to address flexibility, low power consumption and high performance requirements. The presentation will highlight ADAS applications like 360o top view cameras, Camera based automotive mirror replacement, massive CNN processing and the chip reference board demonstration of the final system.

Biography
Jens Benndorf has a PhD. from Moscow University (1991) where he researched DSP technologies for use in high speed voice band modems. Jens’ passion for DSPs continued within international blue chip companies including Alcatel, BridgeCo, Infineon and Silicon Image. He built and led teams developing products for early DSL and complex imaging applications. In 2010 Jens cofounded DreamChip Technologies GmbH and now leads the company in his role as MD and COO. DreamChip specialise with 80 engineers team in multi processor system on chip designs for automotive camera and imaging applications. The company is based in Hannover and Hamburg, Germany and is a member of the Silicon Saxony network.

Wafer Foundry Solutions for Smart Mobility Systems

Alexander Muffler
Business Line Manager Automotive
X-FAB Semiconductor Foundries AG

Abstract
This presentation will show the challenges seen from the viewpoint of a silicon foundry for analog/mixed signal devices developed and manufactured for Smart Mobility Systems. Smart Mobility Solutions require a wide range of semiconductor solutions. In addition to the main processing units and software solutions multiple sensor interfaces and actuator drivers are required in Smart Mobility Systems which are the interface to the analog world. These interface chips need to be designed on robust and highly reliable semiconductor processes. Such devices have to cope with the harsh automotive environments and often have to include high voltage, non-volatile memory (NVM) and extended temperature ranges on one chip. This needs to be considered not only already by the foundry at semiconductor process and NVM IP development stage but also during manufacturing in automotive certified semiconductor factories. To enable IC designers developing such high reliable products specific foundry support with simulation models which behave as close as possible to the real silicon, but also tools for lifetime calculation based on user defined mission profiles are available as well as memories with built-in ECC are just a few examples what the foundry has to offer in addition to a very close collaboration between the foundry and its clients to ensure highest possible quality with zero incidents during lifetime.

Biography
As Business Line Manager Automotive, Alexander Muffler is responsible for the development of product marketing strategies for Automotive semiconductor processes as well as NVM IP for integrated circuit technologies. Prior to joining X-FAB, he worked for Chartered Semiconductor as EDA Manager Europe. Earlier, Alexander worked as an ASIC design engineer at Thesys GmbH. Alexander has more than 25 years of semiconductor technology and design experience. He holds a Master’s degree in Telecommunication Engineering from the University of Applied Science in Konstanz, Germany.

Challenges in detecting and achieving highly reliable microelectronic products. Support by physical and chemical analysis.

Gerald Dallmann
Division Manager Microelectronics
SGS INSTITUT FRESENIUS GmbH

Abstract
Current microelectronic products achieve failure rates in the low-ppm region. But the requirements for field failure reduction are still increasing as more and more microelectronics products like sensors, actuators and integrated systems are introduced in smart vehicles. Autonomous driving will create a new level of unacceptance of system failures. On the other hand, chip and microelectronic product test and qualification procedures are still similar to the old procedures of the AEC Q100 standard with very limited statistics and use conditions. The talk will show examples of failures not covered by qualification and production tests as well as some approaches to improve reliability assessment process. The physical and chemical analysis methods and tools (like XPS, AES, TOF-SIMS) are highly developed to analyze materials and failure modes in semiconductor and system level technologies. The application of these analysis techniques after climate, voltage or mechanical stress application can create a much deeper view inside chip and system weaknesses and failure modes. The talk shows some typical failure mechanisms found in a lab as service provider for many different companies. Some weak spots are discussed with recommendations for improvements.

Biography
Division Manager at SGS Institut Fresenius GmbH in Dresden, Germany, since 2009. Main focus on material and failure analysis of semiconductor devices of client companies. 1995 Director for technology development at Siemens, Infineon, Qimonda, responsible for process integration, yield enhancement and material and technology development of DRAMs. 1990 Product manager microelectronics at Institut Fresenius in Dresden. Main Focus on failure analysis of semiconductor devices. Department manager electron microscopy at Zentrum Mikroelektronik Dresden (ZMD). 1986 Diploma in Microelectronics Technology and Semiconductor Devices.