Schneider, Daniel
Refine
Document Type
Language
- English (6)
Publication reviewed
- begutachtet (6)
Keywords
- Automated Driving (2)
- Ground Truth MB (2)
- advanced driver assistance systems (2)
- automated driving (2)
- Autoamted Driving (1)
- Big Data (1)
- Data Analysis (1)
- Fieldbus (1)
- FlexRay (1)
- Hardware-in-the-Loop (1)
As the development of advanced driver assistance systems (ADAS) continues, more and more software functions and sensors are being introduced to the market. This is accompanied by an increase in the amount of data that has to be transmitted to multiple receivers in the vehicle under hard real-time requirements. The use of deterministic and non-deterministic Fieldbus protocols enables communication between sensor and actuator or ECUs. For the purpose of verifying and validating the developed software modules, but also for type approval, an objective and thus data-driven toolchain is mandatory. By using suitable middleware such as Robotic Operating System (ROS), the complexity of integrating multiple (reference) sensors as well as prototypical software functions can be broken down into subtasks and thus distributed to the hardware in a computationally efficient manner. Recording and manipulating sensor ECU communication while driving is also possible under certain circumstances. However, at least to our knowledge, there is no public ROS driver available to integrate automotive-specific fieldbus protocols except for CAN. In the following paper, we introduce a generic and open-source framework for integrating on-board communication of various Fieldbus protocols and demonstrate the integration in ROS as a real-world use case. To validate the presented methodology, we perform a time analysis of the presented ROS node and compare it to a ROS-independent reference measurement system while performing a standardized vehicle dynamic driving test. In addition, we objectively compare two different on-board sensors from a series vehicle with two distinct reference sensors in a real-world scenario.
In recent years, new challenges have emerged in the automotive sector, particularly in the ADAS/AD domain. The development and testing of such functionality require not only efficient processing and analysis of a rapidly growing amount of recorded vehicle data, but also the ability to deal with a diverse set of new sensor and data types. With its ADAS/AD Big Data & Analytics Platform, AVL addresses these challenges and of-fers a solution that enables highly efficient and scalable search, visualization and anal-ysis of large data sets in an integrated way. A key aspect here is the abstraction of the involved big data mechanisms, since typical users are domain but not big data experts. Additionally, by following standard formats such as Open Simulation Interface (OSI), the ADAS/AD Big Data & Analytics Platform is also broadly applicable. In order to demonstrate the benefits of our platform, we show how it can be applied to logically describe, identify and analyze complex and custom driving scenarios and how tech-nical and legal requirements, such as on an automatically commanded steering function, can be efficiently verified on a large number of test drives.
Today’s validation of driver assistance systems and automated driving functions in state-of-the-art vehicles, still takes place mostly on the proving ground or on the real road. But driving tests can only be performed with the finished vehicle and fully implemented function, which might be difficult in early stages of the development process. Those purely physical tests are also time consuming and lack reproducibility. Hence, the validation process is guided more and more by simulation tools to reduce the number of tests that have to be driven. The pure simulation offers benefits, such as the functions can be tested in early stages of the development without the need for a physical carrier vehicle. Also, the flexibility of the simulation is high, since all variations of driving scenarios, also safety critical, can be driven easily and much faster than real time. On the other side, compared to the physical test, the parameterization of supplemental models is sometimes difficult and the risk of errors due to model abstraction is high.
Due to the rapid progress in the development of automated vehicles over the last decade, their market entry is getting closer. One of the remaining challenges is the safety assessment and type approval of automated vehicles, as conventional testing in the real world would involve an unmanageable mileage. Scenario-based testing using simulation is a promising candidate for overcoming this approval trap. Although the research community has recognized the importance of safeguarding in recent years, the quality of simulation models is rarely taken into account. Without investigating the errors and uncertainties of models, virtual statements about vehicle safety are meaningless. This paper describes a whole process combining model validation and safety assessment. It is demonstrated by means of an actual type-approval regulation that deals with the safety assessment of lane-keeping systems. Based on a thorough analysis of the current state-of-the-art, this paper introduces two approaches for selecting test scenarios. While the model validation scenarios are planned from scratch and focus on scenario coverage, the type-approval scenarios are extracted from measurement data based on a data-driven pipeline. The deviations between lane-keeping behavior in the real and virtual world are quantified using a statistical validation metric. They are then modeled using a regression technique and inferred from the validation experiments to the unseen virtual type-approval scenarios. Finally, this paper examines safety-critical lane crossings, taking into account the modeling errors. It demonstrates the potential of the virtual-based safeguarding process using exemplary simulations and real driving tests.
The focus of this publication is on the development of lane-precise “Ground Truth” (GT) maps for the objective quality evaluation of automated driving functions. Therefore, attention is paid to the proper measurement of road geometry. The road geometry forms the basic layer of the HD maps.
A new map format Curved Regular Objects (CRO) is developed, which is based on the idea of OpenCRG®. For the evaluation of current Advanced Driver Assistance Systems (ADAS) an accurate High Definition (HD) maps as GT are necessary. This makes it possible to locate the high precision vehicle position and motion with centimeter accuracy. The aim is to achieve maximum accuracy of the absolute 3D positions when measuring lanes. This method for the generation of highly accurate GT maps promises an absolute accuracy of < ± 0.05 m. Various research activities benefit from the exact street reference at the Kempten University of Applied Sciences (UAS Kempten) in the Adrive Living Lab. First of all, the publication deals with the current Lane Keeping Assistant Systems (LKAS). The accuracy of the vehicle’s localization on the GT map and an objective evaluation of the LKAS is shown. In addition, the CRO data is used as a virtual sensor for the steering assistant in real time. Another application is the Visual Range Finder (VRF), which requires less computation power by using the CRO data. In addition, a current LKAS camera sensor performance is evaluated using CRO maps.
Advanced driver assistance systems (ADAS) of longitudinal control are widely used. In contrast to longitudinal controls, lateral controls are a growing market since this technique plays a major role in a successful introduction of automated driving. Customer and benchmark studies conducted by theUniversity of Applied Sciences Kempten and Consline AG have clearly shown that the vehicle behavior and customer experience such as tracking performance, driver-vehicle interaction, availability, degree of stress and the sense of security of today's lane keeping assistance systems are consistently rated as extremely unsatisfactory. As a consequence, there is a moderate level of trust and low customer acceptance. A new measuring method based on high-precision and accurate digital maps (ground truth) was developed. With this method, analysis of the entire chain of action, from sensor to tracking is possible. Position, direction and motion of the vehicle and its reference distance to road markings can be precisely measured in the digital map using a high-precision inertial measurement system (IMU) with RTK-DGPS and SAPOS correction service. The measuring method can be used in particular on public routes, since test areas are still insufficient due to the very small tracks and driving maneuver variations for lane keeping assistance systems. For a precise assessment of the sensor, planning and control performance as well as the overall driving characteristics, a very precise knowledge of the routes and the route excitation is required. For this purpose, high precision and accurate digital maps (ground truth) of real tracks were generated. A roof mounted stereo camera system combined with an RTK-DGPS IMU was used to provide offline-generated digital maps with high precision in the OpenDRIVE or OpenStreetMap format, as well as other common simulation formats like IPG CarMaker. In order to be able to carry out the dynamic driving evaluation as well as the simultaneous evaluation of the sensor, planning and control performance in the digital maps in real time, a route format with a regular grid, based on OpenCRG (Curved Regular Grid), was further developed. An IMU with RTK-DGPS and correction service (e.g. SAPOS) provides in real time the highly accurate position, direction and movement of the ego vehicle of up to two centimeters in the lateral and longitudinal direction. In addition, a special measuring steering wheel was built to objectify the driver-vehicle interaction, in particular the steering torque curve and the tracking. Particular attention was paid to the reuse of the original steering wheel with all functions, such as airbag, operation and hands-off detection. The novelty is the ability to measure the recognition, planning and control performance of environmental sensors, algorithms and controllers compared to the reference "Ground Truth". In addition, the driving characteristics of the entire vehicle can be assessed in terms of its tracking performance, driver-vehicle interaction, availability, degree of relieving and a sense of security. Another novelty is the consistent use of digital maps in driving tests as well as in the MIL / SIL / HIL simulation as a digital twin.