Daimler Center for Automotive Information Technology Innovations

Vehicular traffic congestion is a global phenomenon that has increased in importance in the last decades and has caused economically and ecologically negative effects. Thus, finding a way to improve traffic efficiency is a high-frequented problem to be solved by scientists and politicians worldwide. One new promising approach is the usage of decentralized wireless vehicle to vehicle communication based on the Vehicle-2-X (V2X) technology. The idea is that vehicles share information about the current local traffic situation and use this information to optimize their routes. In this paper, we introduce a new algorithm that can be used by navigation systems to calculate routes circumnavigating congested roads. For this purpose, each vehicle transmits its average speed of a road segment to vehicles in the neighbourhood. As a result, vehicles receiving this information can recalculate their routes based on the knowledge about the current possible speeds in the road segments of their neighbourhood. To evaluate the improvements that can be achieved by our algorithm, simulations have been done. Our results show that navigation systems using the V2X technology for a more intelligent route calculation can improve the traffic efficiency of future transport systems.

Mixed traffic of conventional and automated vehicles entails many challenges concerning the safe coexistence and the improvement of traffic efficiency. As a solution, a Traffic Management Center (TMC) combines information from physical sensors and communicated data to enable advanced traffic controllers. Since there is no sufficient penetration rate of automated vehicles nowadays, such a solution could be developed and tested with synthetically generated data using simulation, which models the future Intelligent Transportation System (ITS). This paper addresses the challenges for evaluation of such a system by the means of simulation. For this purpose, we present the Eclipse MOSAIC Multi-Domain Simulation Framework, which provides the possibility to couple best-in-class simulators of the domains traffic, application, and communication, as well as to include further models and external native code libraries. Moreover, we discuss modeling extensions for traffic control algorithms and our method to create a highway scenario based on real traffic data. As primary result, many research questions for mixed traffic and ITS could be investigated using the MOSAIC framework. In an exemplary study, we analyze the deployment effort for infrastructure and communication technologies in order to facilitate a Mainstream Traffic Flow Controller in a realistic scenario with mixed traffic. Simulations performed with MOSAIC confirmed that traffic efficiency improvement is feasible even with fewer infrastructure components, thus with less cost.

High definition (HD) map data is a key feature to enable highly automated driving. With the advent of highly automated vehicles, car makers and map suppliers investigate new approaches to create and maintain HD maps by using on-board sensor data of series vehicles. While state-of-the-art-approaches focus on position and speed data analysis, the consideration of additional vehicle sensor data allows for novel approaches in the context of HD maps. By 2020, more than 30 million connected vehicles are expected to be sold per year, which will generate millions of terabytes of vehicular probe data. One of the major upcoming research issues is to find methods to exploit that probe data to generate and maintain HD maps. In this paper, we address how to develop such methods. We introduce a scalable infrastructure, which supports the ingestion, management and analysis of huge amounts of probe data. It supports an iterative process to develop, assess and tune methods for generating HD maps from probe data. We present a metric to assess methods regarding resulting map precision. As a proof of concept, we present an approach to derive road geometry of highways from location and sensor information.

Automated vehicles rely on the accurate and robust detection of the drivable area, often classified into free space, road area and lane information. Most current approaches use monocular or stereo cameras to detect these. However, LiDAR sensors are becoming more common and offer unique properties for road area detection such as precision and robustness to weather conditions. We therefore propose two approaches for a pixel-wise semantic binary segmentation of the road area based on a modified U-Net Fully Convolutional Network (FCN) architecture. The first approach UView-Cam employs a single camera image, whereas the second approach UGrid-Fused incorporates a early fusion of LiDAR and camera data into a multi-dimensional occupation grid representation as FCN input. The fusion of camera and LiDAR allows for efficient and robust leverage of individual sensor properties in a single FCN. For the training of UView-Cam, multiple publicly available datasets of street environments are used, while the UGrid-Fused is trained with the KITTI dataset. In the KITTI Road/Lane Detection benchmark, the proposed networks reach a MaxF score of 94.23% and 93.81% respectively. Both approaches achieve realtime performance with a detection rate of about 10 Hz.

Vehicle-2-X (V2X) Communication provides the foundation for new applications that enhance both safety and traffic efficiency. Before V2X applications can be deployed in practice, their in-depth analysis is necessary. For this end, detailed and realistic simulations are essential. Depending on the simulated V2X Communication application, particular simulators have to be coupled. For this purpose, we have developed the V2X Simulation Runtime Infrastructure (VSimRTI) offering the flexibility to combine arbitrary simulators. The VSimRTI is derived from concepts of the High Level Architecture (HLA). It synchronizes the simulators and enables the communication among them. Another feature of our simulation environment is the emulation of the environment of V2X Communication applications in real vehicles. As a result, we can integrate real V2X Communication applications without modifications.

The proposed benefits of enabling automated and autonomous vehicles to cooperate are manifold - however, these functions introduce a new level of uncertainty and unreliability inherent of wireless communication into a realm of safety-critical decisions. Since vehicle-to-vehicle communication in either ad hoc or managed environments can be inherently unreliable, it is of highest importance to critically evaluate the level and design of integration of cooperative information into the decision making process of automated functions. Thus, a robust integration of communication as a sensor has to take into account key issues such as penetration rate, reliability of communication and trust and develop appropriate methods of handling these issues to provide fail-safety. In this paper we present an approach to cooperative maneuvers in automated vehicles with emphasis on handling potential hazards introduced by communication. In this regard we propose the complimentary Collaborative Maneuver Protocol (CMP), combining novel approaches to enable robust, functionally safe collaboration between vehicles in vehicle-tovehicle communication.

Positioning systems play an ever increasing key role in vehicular applications. GNSS-based systems such as GPS represent the predominant primary technology for positioning in this context. However, in densely built-up urban areas, the positional accuracy of GNSS-based systems decreases significantly, and ceases operation indoors due to the lack of line- of-sight to the satellites. In these scenarios and for use cases in which GNSS-based systems do not meet the requirements, the need for alternative localization systems arises. There is a wide range of vehicle indoor positioning approaches ranging from optical systems over IMU-based systems to LiDAR SLAM. In general, all systems can usually be classified by their perspective (internal or external) and their order (absolute or relative). Furthermore, all considered systems are very application-specific, and additionally either require a comprehensive extension of the existing infrastructure or modification of the vehicle.

The introduction of dedicated short-range Vehicle- to-Vehicle communication (DSRC) enables the next step in advanced driver assistance systems (ADAS) - the cooperative automated driver assistance systems (CoDAS). Combined with automated functions and even autonomous driving, a host of novel functions become feasible. Some of these - such as platooning- have been in research for decades, while others are not tackled yet. In this paper we give an overview on research on automated cooperative functions, survey conceivable functions and present a way to classify them.

In the last decade, on-board computers were continuously substituted by sophisticated in-car-infotainment systems due to the ever increasing demand for mobile services and entertainment features. As a consequence, the automotive user interfaces became more complex and a driver is now faced with the time and attention-consuming task of traversing a menu structure in order to execute a desired functionality. Extending an in-car-infotainment system with the ability to trace recurrent user behavior within an intelligent environment opens up new opportunities for user-adaptive and context-aware user interfaces. This paper introduces two different approaches to simplify the task of executing a preferred feature by either personalizing a list of context-dependent shortcuts or by automatically executing regularly used features. Both approaches are evaluated against an in-car-infotainment system that is currently available on the market. Additionally, a fully functional prototype with a real world driving simulator was implemented in order to conduct an user study within our labs.

Advanced Driver Assistance Systems (ADAS) were strong innovation drivers in recent years, towards the enhancement of traffic safety and efficiency. Today’s ADAS adopt an autonomous approach with all instrumentation and intelligence on board of one vehicle. However, to further enhance their benefit, ADAS need to cooperate in the future, using communication technologies. The resulting combination of vehicle automation and cooperation, for instance, enables solving hazardous situations by a coordinated safety intervention on multiple vehicles at the same point in time. Since the complexity of such cooperative ADAS grows with each vehicle involved, very large parameter spaces need to be regarded during their development, which necessitate novel development approaches. In this paper, we present an environment for rapidly prototyping cooperative ADAS based on vehicle simulation. Its underlying approach is either to bring ideas for cooperative ADAS through the prototyping stage towards plausible candidates for further development or to discard them as quickly as possible. This is enabled by an iterative process of refining and assessment. We reconcile the aspects of automation and cooperation in simulation by a tradeoff between precision and scalability. Reducing precise mapping of vehicle dynamics below the limits of driving dynamics enables simulating multiple vehicles at the same time. In order to validate this precision, we also present a method to validate the vehicle dynamics in simulation against real world vehicles.

Modern vehicles are equipped with numerous driver assistance and telematics functions, such as Turn-by-Turn navigation. Most of these systems rely on precise positioning of the vehicle. While Global Navigation Satellite Systems (GNSS) are available outdoors, these systems fail in indoor environments such as a car-park or a tunnel. Alternatively, the vehicle can localize itself with landmark-based positioning and internal car sensors, yet this is not only costly but also requires precise knowledge of the enclosed area. Instead, our approach is to use infrastructure-based positioning. Here, we utilize off-the shelf cameras mounted in the car-park and Vehicle-to-Infrastructure Communication to allow all vehicles to obtain an indoor position given from an infrastructure-based localization service. Our approach uses a Convolutional Neural Network (CNN) with Deep Learning to identify and localize vehicles in a car-park. We thus enable position-based Driver Assistance Systems (DAS) and telematics in an underground facility. We compare the novel Deep Learning classifier to a conventional classifier using Haar-like features.

The increasing importance of embedded software has produced a shift in the testing activities from system testing towards software testing. This has contributed to testing the core system functionality earlier on in the test process. However, this shift has also led to very similar test cases being both described and executed independently at different test levels. We propose reusing multi-level test cases for supporting seamless test level integration. As a consequence, single test case specifications and implementations are reused throughout the test process, minimizing the test implementation effort and taking advantage of the synergies among test levels.

Evolutionary Functional Testing (EFT) is a relatively recent approach to automating the testing process. The research presented in this paper aims at increasing the acceptability of EFT in industrial settings. An approach suitable for efficiently and effectively testing complex continuous control systems is introduced. The main focus is on generating realistic test stimuli, enabling interactivity between test driver and test object, and facilitating the process of designing a suitable fitness function. This is accomplished by integrating EFT with model-based testing methodologies resulting in an intuitive testing approach that enables even testers not familiar with search based testing to achieve good results with limited effort. A test environment optimized for deployment in the industrial domain is introduced. Features of the test environment include the capability of automatically generating realistic continuous test data sets, interacting with the system under test during test execution, and automatically executing and evaluating large numbers of tests. A thorough case study using an adaptive cruise control system from the automotive domain is performed to assess the approach. Results of this work indicate high usability, efficiency, and effectiveness of the proposed method for testing complex embedded systems.

For the simulation of all aspects of V2X Communication scenarios, different simulators have to be combined and an interaction among them at runtime of the simulation has to be enabled. Hence, we have developed the V2X Simulation Runtime Infrastructure (VSimRTI) which couples discrete event-based sim

Evolutionary testing has been researched and promising results have been presented. However, evolutionary testing has remained predominately a research-based activity not practiced within industry. Although attempts have been made, such as Daimler's Evolutionary Structural Test (EST) prototype, until now, no such tool has been suitable for industrial adoption. The European project EvoTest (IST-33472) team has been working from 2006 till 2009 to improve this situation. This paper describes the final version of the Evolutionary Testing Framework (ETF) resulting from the EvoTest project. In specific we will present the EvoTest Structural Testing tool for fully automatic structural testing that has been demonstrated to be suitable within an industrial setting. The paper concentrates on how to use it and interpret the results. The paper starts with introducing the concepts of Evolutionary Testing in general and Structural Testing in specific. Subsequently, the ETF and the EvoTest Structural Testing tool built on-top of it will be described. We will concentrate on the usage, the architecture, and remaining limitations of the tool. The paper concludes describing the results of using the EvoTest Structural Testing tool in practice on real-world systems in an industrial setting.

In this paper we present an indoor micronavigation system for enclosed parking garages. It builds on car-to-infrastructure communication to provide layout information of the car park, the coordinates of the destination parking lot, as well as external positioning information to vehicles. In our approach we use customary network video cameras to detect and locate vehicles within the car park. Once a vehicle is detected, the system correlates the position of the vehicle to the car park layout and transmits this information to the appropriate vehicle to substitute the internal positioning system. With this information the vehicle is guided from the car park entrance to a destination parking lot.

Test case generation constitutes a critical activity in software testing that is cost-intensive, time-consuming and error-prone when done manually. Hence, an automation of this process is required. One automation approach is search-based testing for which the task of generating test data is transformed into an optimization problem which is solved using metaheuristic search techniques. However, only little work has been done so far applying search-based testing techniques to systems that depend on continuous input signals. This paper proposes two novel approaches to generating input signals from within search-based testing techniques for continuous systems. These approaches are then shown to be very effective when experimentally applied to the problem of approximating a set of realistic signals.

Hundreds of thousands of pedestrians are involved in severe traffic accidents every year world-wide. Reasons for these accidents include complex and highly dynamic traffic situations where views are obstructed or unexpected movement occurs. Driver assistance systems are a valid option for increasing pedestrian safety by enhancing the awareness of complex traffic situations and identifying potential dangers. In this work, we present a collision avoidance system based on smart video surveillance and car-to-infrastructure communication. We use a distributed system of monocular cameras to determine the position of both vehicles and pedestrians in realtime. In addition, we utilize standard car-2-x communication technology (ETSI ITS G5) to provide all position detections to the vehicles, thus enabling complex use cases such as warning cascades to drivers in case of oncoming dangers. A detailed evaluation of the proposed system and collision warning use case demonstrates the suitability as assistance system for human drivers. We also show that automatic braking systems would lead to drastic performance improvements due to a significant reduction of reaction times.

Model-based software design is constantly becoming more important and thus requiring systematic model testing. Test case generation constitutes a critical activity that is cost-intensive, time-consuming and error-prone when done manually. Hence, an automation of this process is required. One automation approach is search-based testing for which the task of generating test data is transformed into an optimization problem which is solved using metaheuristic search techniques. However, only little work has been done so far applying search-based testing techniques to continuous functional models, such as SIMULINK STATEFLOW models. This paper presents the current state of my thesis developing a new approach for automatically generating continuous test data sets achieving high structural model coverage for SIMULINK models containing STATEFLOW diagrams using search-based testing. The expected contribution of this work is to demonstrate how search-based testing techniques can be applied successfully to continuous functional models and how to cope with the arising problems such as generating and optimizing continuous signals, covering structural model elements and dealing with the complexity of the models.

Cooperative Adaptive Cruise Control (CACC) is considered as a key potential enabler to improve driving safety and traffic efficiency. It allows for automated vehicle following using wireless communication in addition to onboard sensors. To achieve string stability in CACC platoons, constant time gap (CTG) spacing policies have prevailed in research; namely, vehicle interspacing grows with the speed. While constant distance gap (CDG) spacing policies provide superior potential to increase traffic capacity than CTG, their major drawbacks are a smaller safety margin at high velocities and that string stability cannot be achieved using a one-vehicle look-ahead communication. In this work, we propose to apply CDG only in a few driving situations, when traffic throughput is of highest importance and safety requirements can be met due to relatively low velocities. As the most relevant situations where CDG could be applied, we identify starting platoons at signalized intersections. With this application scenario we show that applying CDG only in a few specific and crucial situation can have a major impact on traffic efficiency. Specifically, we compare CTG with CDG regarding its potential to increase the capacity of traffic lights. Starting with the elementary situation of single traffic lights we expand our scope to whole traffic networks including several thousand vehicles in simulation. Using real world data to calibrate and validate vehicle dynamics simulation and traffic simulation, the study discusses the most relevant working parameters of CDG, CTG, and the traffic system in which both are applied.