Advanced Microsystems for Automotive Applications 2009

EE-VERT is a project funded under the Seventh Framework Programme of the European Commission concerned with improving the energy efficiency of road vehicles. In particular EE-VERT targets a 10% reduction in fuel consumption and CO

2

generation as measured on the standard NEDC. The central concept to EE-VERT is the deployment of an overall energy management strategy that coordinates energy demand and consumption by a network of smart generators and smart actuators. This paper describes the EE-VERT concept and the activities that will be undertaken to realise the project’s goals.

Future Electric (EVs) and Hybrid Electric Vehicles (HEVs) will provide more flexibility when choosing between primary energy sources, including those which are renewable. In general conventional ICEs vehicles transform between only 17 and 22% (depending on power train) of the fuel chemical energy with a typical primary energy consumption of 550-600 Wh/km (0.06 l/km). Efficient electrically powered trains can achieve conversion efficiencies greater than 75% from the batteries to the wheels, which corresponds to consumption in primary energy of about 390 Wh/km in the case where electricity is produced by conventional carbon based power plants, or only 180 Wh/km where the electricity is produced solely by renewable energy. The partial recovery of kinetic energy during braking gives rise to further improvement in the overall efficiency. The development of advanced smart electronic systems in power trains is therefore essential for delivering a considerable energy saving in terms of the most critical sources (oil and natural gas - NG). This paper presents the advances made in the overall power electronic modules for electric and hybrid vehicles, and which are addressed in the E

3

Car project.

The aim is an additional sensor system that is able to detect left behind occupants in parked cars. This should avoid the decrease of fatalities found in unattended or oversight individuals in vehicles. Based on acceleration measurements directly at the car chassis information about the occupancy is extracted. At the beginning of this paper the theory of the signal source, the used car model and the applied classification algorithms is shortly given. Afterwards several measurement results are presented and become analyzed with regard to a following automatic classification. The next step is the evaluation of different classification algorithms and the explanation of the performance on the acceleration datasets that were collected during this research. Many different classification algorithms are available, at this point the support vector machine (SVM), k-nearestneighbor (k-NN), probabilistic neural network (PNN), decision tree and clustering were observed.

The European IP APROSYS (Advanced PROtection SYStems) is currently being finished. The overall project goal was to decrease the number of fatalities on European roads. Within Sub-Project 6 a novel pre-crash side impact protection system was developed which improves structural behaviour in side crash. The main focus is to reduce intrusions. The given paper presents research activities which were carried out within SP6. The pre-crash-system combines a sensor unit, a data processing or data fusion unit and at least one reversible high-speed actuator. As outcome of this study about the pre-crash-system, a crash load redirection to the unstruck side was found to be most powerful and a suitable actuator was developed which takes away the crash loads directly from the incoming object at the door. This was achieved by creating a rigid connection from the struck door to other stiff car regions. By this, the energy absorbing process starts earlier and involves more structural parts. This system changes the crash deformation modes completely. Both, the B-pillar as well as the door intrusions are being significantly reduced, especially in regions being most critical for the occupant.

In this paper a study on the feasibility of a bi static radar system for bad road conditions, including water, snow and ice, is presented. First of all, the most suitable frequency is chosen among all the frequencies available for such a system from a regulatory point of view. Then, the best mathematical model for predicting the electromagnetic behaviour of different layers of water, snow or ice over an asphalt basis is presented. The numerical study is focused on the reflexion attenuation (the difference between the direct and the reflected signal levels). Results for linearly polarized waves are presented and some useful design guidelines are presented.

Today most so called ‘black spots’ have been eliminated from the road networks. However, intersections can still be regarded as black spots. Depending on the region and country, from 30% to 60% of all injury accidents and up to one third of the fatalities occur at intersections. This is due mainly to the fact that accident scenarios at intersections are among the most complex ones, since different categories of road user interact in these limited areas with crossing trajectories.

Traffic of heavy goods vehicles is an important component of transport in today’s cities. Much progress has been made to make it as safe and efficient as possible. In this paper, safety of heavy goods traffic at intersections is investigated. Special regard is given to the safety of vulnerable road users (VRUs), such as pedestrians and cyclist. In the project INTERSAFE-2, solutions are being developed to reduce accident risks at intersections. One situation identified to be particularly dangerous for VRUs is the right turning of the heavy goods vehicle. This scenario will be specifically addressed by he intersection safety application developed on the Volvo Technology demonstrator vehicle in INTERSAFE-2. This paper describes the requirements for resolving the right-turning accident scenario, and will illustrate the resulting function concept of the demonstrator.

Vehicle-to-vehicle and vehicle-to-infrastructure communication (V2X communication) has great potential to increase road and passenger safety, and has been considered an important part of future Intelligent Transportation Systems (ITS). Several R&D projects around the world have been investigating various aspects of V2X communication. Some of these projects focus on specific issues of V2X communication for intersection safety (communication-based intersection safety) because intersections are the most complex driving environments where injury and fatal accidents occur frequently. In this paper, we discuss the technical details of V2X communication and discuss how it can be used to improve intersection safety.

This paper presents experimental results regarding road condition monitoring by machine vision techniques. The system will be further developed by VTT in the European INTERSAFE-2 project, focused on applications at road intersections. Knowing the presence of adverse conditions such as icy or wet roads, important effects can be obtained for intersection safety by means of more effective driver assistance functions. A very good reliability up to 93% has been found for the detection of icy roads, while lower values around 60% have been measured in the case of wet surfaces, due to a higher sensitivity to environmental conditions, especially outdoor light. Future steps will include an advanced classification algorithm and the implementation of active lighting. Integration of the vision equipment into a cooperative system for intersection safety is being investigated.

The intersection scenario imposes radical changes in the physical setup and in the processing algorithms of a stereo sensor. Due to the need for a wider field of view, which comes with distortions and reduced depth accuracy, increased accuracy in calibration and dense stereo reconstruction is required. The stereo matching process has to be performed on rectified images, by a dedicated stereo board, to free processor time for the high-level algorithms. In order to cope with the complex nature of the intersection, the proposed solution perceives the environment in two modes: a structured approach, for the scenarios where the road geometry is estimated from lane delimiters, and an unstructured approach, where the road geometry is estimated from elevation maps. The structured mode provides the parameters of the lane, and the position, size, speed and class of the static and dynamic objects, while the unstructured mode provides an occupancy grid having the cells labeled as free space, obstacle areas, curbs and isles.

In the integrated research project SAFESPOT the concept of a Cooperative Pre-Data-Fusion has been implemented as a novel approach for an environmental perception system. This paper describes the approach of eliminating measurements of a Laserscanner which are not part of the environment of interest. It is being developed in the subproject INFRASENS which specifies and develops an infrastructure-based sensing platform.

Nowadays, most so called ‘accident black spots’ have been eliminated from the road networks. However, intersections can still be regarded as accident black spots. Depending on the region and country, from 30% to 60% of all injury accidents and up to one third of the fatalities occur at intersections. Thus, the European R&D project INTERSAFE-2 aims to develop and demonstrate a co-operative intersection safety system, in order to significantly reduce injury and fatal accidents at intersections. In line with this project, the user needs for intersection safety systems and the users’ impressions respectively attitudes towards such systems were determined with the aid of a questionnaire which has been executed in six European countries. The results of the Europe-wide questioning are presented within this paper.

Partly automated driving is relevant for solving mobility problems, but also cause concerns with respect to driver’s reliability in task performance. The presented supported driver model is therefore intended to answer in which circumstances, what type of support enhances the driver’s ability to control the vehicle. It became apparent that prerequisites for performing tasks differ per driving task’s type and require different support. The possible support for each driving task’s type has been combined with support-types to reduce the error causations from each different performance level (i.e. knowledge-based, rule-based and skill-based performance). The allocation of support in relation to performance level and driving task’s type resulted in a supported driver model and this model relates the requested circumstances to appropriate support types. Among three tested ADAS systems, semi-automated parking showed best allocation of support; converting the demanding parallel parking task into a rather routine-like operation.

Vision-based driver assistance systems are contributing to traffic safety and comfort in a variety of applications. However, due to the limited resolution provided by state-of-the-art cameras, the observable area is very limited. Thus, high resolution image sensors have been developed recently for the automotive area. In this paper, the potential of such camera systems is analyzed on the example of a lane recognition system. In particular, the possible augmentation of the observable area is discussed. However, it is also shown which algorithmic adaptations are necessary in order to benefit from such sensors. For that, a new lane model is proposed which consists of several continuous segments and, thus, is also suitable for higher detection ranges. Finally, first results from a practical implementation of this model are presented and discussed.

Road construction sites on highways are a demanding environment for drivers as the lane width is reduced. Especially for trucks, lateral control is a challenge. If the driver slightly drives over the lane markings, other vehicles cannot use the neighbouring lane. This leads to a reduction of the road capacity, thus causing traffic jams. A driver assistance system which supports the driver in the task of lateral control highly benefits from the use of accurate infrastructure (feature) maps. This paper presents an approach for the automatic detection, classification, and mapping of specific elements of road works like guard rails, safety barriers, traffic pylons, and beacons with laser scanners.

Motorcyclists and moped drivers are road users with a particularly high accident risk since motorcycle accidents are severe in nature, due to the relative minor protection of motorcyclists. Furthermore, today the field of driver assistance systems is mostly dedicated to the passenger vehicle and heavy goods vehicle sector. Driver assistance systems for Powered Two Wheelers (PTWs), which rely on onboard vehicle sensors, have not been not considered so far. In this paper we present first results of the European funded Marie Curie research project MYMOSA and especially the Integrated Safety work within its work plan. The objective of the Integrated Safety work package is the development of an integrated safety system capable to detect impending dangerous situations and accident scenarios. As a consequence of the proposed activities in these projects a long-term reduction of at least 20% of injuries and fatalities of motorcyclists is foreseen.

In a future car-IT architecture consisting of a centralized grid-cluster of good-performance electronic control units connected via a switched high-bandwidth communication network, new possibilities for driver assistance systems come up. Working on raw data in combination with strong central processing units allows advanced signal processing. In case of breakdown which results in reduced available communication bandwidth and/or reduced available calculating power, graceful degradation will help to keep the system running. That means either the data acquisition can work at lower performance (lower angular resolution, lower update rate etc.) to reduce the calculation power demand as well as the network data rate, or a part of the signal chain can dynamically be shifted to smart-sensors and make them work similar to current sensors like Bosch’s Adaptive Cruise Control [1]. In this case, only a list of objects instead of raw data is transmitted to the electronic control unit where a (brake- or accelerate-) decision could be made with low calculation power while accepting a lower overall system performance.

The paper represents a vision-based lane departure warning system, as well as, a driver assistance system for the automatic recognition of traffic signs. It could inform the driver about actual speed limits or passing restrictions, which could be ignored in difficult traffic situations. The system is implemented partially on a new multi-core processor, developed by the Infineon Technologies AG.

This paper presents the test-bed that will be developed for a Unified Perception & Decision Architecture (UPDA). Due to the increasing demand of ADAS systems to be mounted on cars, it is more and more important to develop a unified architecture that can communicate and share information between these systems. This is the aim of an ERC-founded project and to develop and test such architecture a car has been set up with many different sensors.

With the increasing demand on driver information and multimedia content, displays and video signalling are getting more and more attention in the automotive industry. The requirement of transmitting video signals includes applications like infotainment displays, dashboard and head-up displays, but also driver assistance systems requiring real time video streams. The Automotive Pixel Link (APIX) Technology has been specifically designed to address the different requirements for video transmission in automotive applications. Optimized for low EMI, the APIX technology offers the ability to combine high speed real-time video data, full-duplex sideband channels and the power supply over a single cable. Especially on multicamera applications, the video information needs to be provided with minimum delay, with special requirements on synchronization and image processing. This article provides an introduction on the APIX technology, followed by the discussion of various requirements of implementing camera based driver assistance systems in the automotive environment.

Tire pressure monitoring systems (TPMS) are beneficial for the environment and road and passenger safety. Miniaturizing the TPMS allows sensing of additional parameters. This paper presents a miniaturized TPMS with a volume less than 1 cm

3

, realised by 3D stacking and through-silicon via (TSV) technology. Suitable technologies with low electrical resistance and high bond strengths were evaluated for stacking the microcontroller, transceiver, pressure sensor and bulk acoustic resonator (BAR) in the TPMS. 60 μm deep W-filled TSVs with resistance 0.45 Ω and SnAg micro bumps with a bond strength of 53 MPa were used for stacking the transceiver to the microcontroller. TSVs through the whole wafer thickness with resistance 6 Ω were used for the pressure sensor. Au stud bumps were used for stacking the pressure sensor and BAR devices. The final TPMS stack was packaged in a moulded interconnect device (MID) package.

MEMS (Micro-Electro-Mechanical Systems) are being included in an increasing number of applications in automotive systems, particularly because of their small size and cost. However, most automotive applications require more information than can be provided by a single MEMS motion sensor. We report on the development of a MEMS motion sensor that can detect acceleration along three axes and angular velocity around two or three axes using a single proof mass, enabling the integration of multiple applications into a single device. The sensor consists of four support springs patterned into the device layer of an SOI (Silicon on Insulator) wafer and a single proof mass formed by etching through the handle layer using DRIE (Deep Reactive Ion Etching). We present the fabrication and packaging process and modeling results for the sensor.

With the launch of the AIS326DQ, its first AECQ100-qualified 3-axis accelerometer, STMicroelectronics is addressing non-safety applications like vehicle alarms, tracking and monitoring, black-box systems, and navigation assistance. Government regulations, insurance requirements, and a general increasing user demand for car security are the main driving factors of this high-CAGR market. Fully leveraging the economies of scale of a proven 200 mm wafer technology (STMicroelectronics was the first company in the world to start MEMS volume production on 8” wafers), advanced features such as user selectable full scale ranges, 12-bit usable resolution, together with the mounting freedom provided by the 3-axis sensing, make AIS326DQ the recommended choice when customers look for the best trade-off between performances and price. STMicroelectronics will continually be enriching its offer by complementing the portfolio with more innovative solutions: the new generation of accelerometers is reported to show a state-of-the-art thermal stability below 0.2 mg/°C, while, after a successful launch in the consumer segment, ST gyroscopes will find their way into automotive applications.

Fault-tolerant devices are becoming more and more important in safety-critical applications. In addition, because of further decreased geometries, integrated circuits are becoming more susceptible to induced interference. This paper presents new methods and design concepts to make application specific integrated circuit (ASIC) devices fault-tolerant to effects generated in the harsh automotive environment, especially to single event effects (SEEs). We describe how to mitigate single event effects which can immediately affect the function of electronic components. ASICs provided with this technique will increase the reliability and dependability while simultaneously maintaining the full real-time behaviour of the system.

Digital sensors offer the possibility to increase globally the functionality of automotive systems and at the same time to further reduce their cost. The Single Edge Nibble Transmission (SENT) digital communication protocol is a promising low-cost solution for communication between sensor satellites and a microcontroller. This paper presents the Short PWM Code (SPC) protocol aiming at extending the SENT communication link in terms of functionality, performance and cost efficiency. SPC provides application relevant functionalities which are not included in SENT while still keeping as close as possible to the original protocol. In particular: bi-directionality, synchronicity and bus capability.

Sensor data fusion of active near infrared (NIR) and passive far infrared (FIR) for reliable detection of vulnerable road users in future warning automotive night vision systems requires for low-cost, mid-resolution FIR sensor arrays for hot spot detection. We present a new cost efficient technology for FIR arrays adopting a volume proven integrated MEMS process for the production of a suspended thermo-diode array. In contrast to established bolometer production all steps of the process developed are fully semiconductor compatible as the sensor element formation is an integral part of the read out IC processing and does not require ASIC backend processing with dedicated equipment. Vacuum wafer-level packaging compatibility further reduces cost. In a first step the proposed process has been verified with small integrated FIR arrays consisting of 42x28 pixels. The FIR array development reported is part of the EU FP7 project ‘ADOSE’.

Future automobiles will be faced with many constraints based on environmental and safety legislation. In addition further demands such as cost reduction, differing fuel types and improved driver comfort will be made. At first view many of these demands are conflicting. This paper describes the role that sensors and in particular the interface Application Specific Integrated Circuits (ASIC) or specialised Application Specific Standard Products (ASSP), play in meeting these exacting requirements. The latest advances in sensing element technology must be matched with an equivalent improvement in the performance of the readout electronics to produce the complete sensor package. Requirements of several sensing elements are reviewed and how the ASIC design responds to these challenges will be explained. Examples in the fields of power train efficiency using precise angle measurement and pressure sensors will be given. Other applications such as maintaining maximum night time driver vision and the monitoring of fuel sources for maximum efficiency will be outlined.

Airbag electronic systems measure various sensor signals and control several different actuators in an effort to prevent passenger injuries in case of accidents. This paper describes the electronics needed in such an application (acceleration sensors, bus interfaces, uC, power supply, squib drivers etc.) and the development of airbag systems as well as their increase of complexity over time. Acceleration information is delivered via PAS, PSI5 or DSI busses by high g acceleration sensors installed on vehicle corners and pressure sensors installed inside the doors. This information is validated by the microcontroller vs. centrally located inertial sensors. Precisely sampled acceleration values are used to identify a crash situation, in which case the firing circuit is enabled and a firing signal is sent to deploy the airbags. Building blocks like voltage regulator/power supplies and squib drivers have to provide the energy at the right time to ignite the airbag balloons mounted in the front and sides of the vehicle and tighten seat belts. Over time, new features have been added to the system - including rollover protection with airbag curtains, rear crash protection (whiplash prevention), and superior advanced crash detection algorithms to fire multi-stage airbags. All this functionality needs to be addressed by effective, safe, reliable, cost-effective electronics. Integration of the functions in a few application specific ICs and the fact that all function blocks and necessary semiconductor processes are available in one company makes the creation of synergies and the implementation of cost-effective silicon strategies possible.

The detection of the omni-present event chains in embedded applications goes far beyond functional modelling and static analysis. Once identified the analysis of their dynamics reveals a lot of data about the system like stability, critical paths or load reserves for future extensions. By using task-models and a real-time simulation tool the detection and analysis of event chains is very easy. Especially in distributed and collaborative development environments this is very helpful in reaching not only functional perfect systems but also delivering a high level of real-time quality.

After supplying high-performance low-g accelerometers to automotive systems for 10 years, it was time to broaden the product scope in order to utilize VTIs 3D MEMS platform for more products. Based on a combination of MEMS-tools that is effectively used in low-g high-volume production, a gyroscope and accelerometer toolbox was set-up in order to address both worlds, consumer and automotivesafety markets. The requirements of the consumer and automotive worlds are often contradictory, and synergies difficult to achieve. It can be agreed for the functional design that it needs to follow the applicational requirements – e.g. low power for consumer vs. strong and robust signal for automotive. It will be demonstrated that there are also commonalities. Regarding automotive system integration, the controversial discussion continues since functional and environmental aspects dictated by the application or by sensor fusion often lead to complex requirements. The art of bundling is to draw a line between combinable sensor classes and classes where a derivate of the sensor will fulfill deviating needs, which could be as simple as sensing orientation, temperature and vibration environmental conditions as well as signal quality requirements.

Despite the fact of permanently increasing road traffic, the number and the severity of accidents have been reduced thanks to passive and active safety systems. These systems protect road users before and during traffic accidents. Driver assistance systems inform the driver about potentially dangerous situations and intend to bring him back into the loop. The key to reliable warning and intervention lies in the performance of the sensor system. Reliable object information requires the development of intelligent sensor systems. This article describes the architecture for an adaptive radar sensor subsystem. The architecture allows the configuration of the radar sensors dependent on the traffic situation. For this purpose, the bandwidth of the radar sensors is adjusted by means of ego vehicle speed and information about tracked objects (position and speed). Furthermore the approach improves the radar signal processing by searching objects in the radar spectrum.