Logo: to the web site of Uppsala University

Autonomous driving is one of the key technologies for increasing road safetyand reducing traffic volumes. Therefore, science and industry are workingtogether on new innovative solutions in this field of technology. One importantcomponent in this context is the approval and testing of new solution concepts,with special focus on the ones for urban environments. Not only because ofthe high diversity of traffic situations, but also because of the close contactbetween vulnerable road users (VRU) and automated vehicles.In the course of this work, a novel approach for testing automated drivingfunctions and vehicle systems in urban environments is presented. The goal isto create a safe and valid environment in which the automated vehicle and theVRU can meet and interact. The basis is a highly realistic virtual model of acity center. The physical behavior of the vehicle and VRU is recorded usingmeasurement technology and transferred to the virtual city model.Based on representative urban traffic scenarios, the functionality of the urbantest field is investigated from various points of view. Thereby, the focus is onreal-time capability and the quality of interaction between the vehicle and theVRU.The investigations show that both the real-time capability and the interactionpossibilities could be demonstrated. Further, the developed methodologies aresuitable for real time applications.

Advanced driver assistance systems are an important step on the way towards the autonomous driving. However, there are new challenges in the release of increasingly complex systems. For the testing of those systems many test kilometers are necessary to represent sufficient diversity. Hence, the virtual testing of driver assistance systems brings new opportunities. In virtual environments, it is possible to run a much higher distance in a short time. Simultaneously, the complexity of the environment and the test scenarios are individually adjustable. It is possible to test scenarios that are not feasible in a real environment due to a risk of injury. A big challenge is the physical correct implementation of real vehicles and their components into the Virtual Reality. To enable a realistic virtual testing the vehicles surrounding sensors need to be modeled adequately. Thus, this paper presents an approach for the implementation of a Lidar model into a Virtual Reality. A physical Lidar model is combined with a real-time capable vehicle dynamics model to investigate the influence of vehicle movements to the sensor measurements. The models are implemented into a highly realistic virtual city environment. Finally, a test campaign shows the influence of the Lidars physics and the vehicle dynamics on the detection results.

New approaches for testing of autonomous driving functions are using VirtualReality (VR) to analyze the behavior of automated vehicles in variousscenarios. The real time simulation of the environment sensors is still a challenge.In this paper, the conception, development and validation of an automotiveradar raw data sensor model is shown. For the implementation, theUnreal VR engine developed by Epic Games is used. The model consists of asending antenna, a propagation and a receiving antenna model. The microwavefield propagation is simulated by a raytracing approach. It uses the methodof shooting and bouncing rays to cover the field. A diffused scatteringmodel is implemented to simulate the influence of rough structures on thereflection of rays. To parameterize the model, simple reflectors are used. Thevalidation is done by a comparison of the measured radar patterns of pedestriansand cyclists with simulated values. The outcome is that the developedmodel shows valid results, even if it still has deficits in the context of performance.It shows that the bouncing of diffuse scattered field can only be doneonce. This produces inadequacies in some scenarios. In summary, the papershows a high potential for real time simulation of radar sensors by using raytracing in a virtual reality.

Recently, virtual realities and simulations play important roles in the development of urban traffic infrastructure. By anappropriate abstraction, they help to design, investigate and communicate inner-city development processes. Especially, to investigate interactions between infrastructure and future mobility participants, a valid virtual model is essential for functionality and reliability.

The aim of this study is the investigation of interactions between a virtual infrastructure model and virtual sensor models of highly automated mobility systems. The overall system consists of a georeferenced virtual city model and probabilistic sensor models, which are part of an automated vehicle model. The virtual environment comprises a comprehensive, virtual 3D model of a representative German inner-city scene, considering specific height coordinates. The probabilistic sensor models represent real radar and lidar sensors and comprehensively replicate their physical functionality in a virtual environment. Considering different levels of detail, the realistic representation of physical effects of the virtual city model on the virtual sensors is investigated. The investigated scenarios are derived from representative urban traffic situations. The complexity as well as the level of information of the virtual city scenarios is iteratively increased. Subsequently, the effects on the model validity of the overall system is checked and analysed.

The results show that the developed virtual environment performs well for different levels of detail of representative virtual traffic scenes. In addition, the selected modelling depth is very suitable for the high-performance investigation of interaction between virtual urban environment and virtual autonomous vehicle.

For many years now, models for representing reality have played a decisiverole in the development of control systems. By appropriate abstraction theyhelp to design an efficient development process. Especially in the developmentof Advanced Driver Assistance Systems (ADAS) a valid virtual developmentenvironment is crucial for functionality and reliability.

This study aims the representation of driving physics in a virtual testenvironment for the development of robust ADAS systems. The overall systemconsists of a georeferenced virtual traffic environment, a multibody vehiclemodel and a driver model. The virtual environment includes a detailed 3D model of an urban city in consideration of specific height coordinates of theenvironment. The vehicle model is implemented by a simplified two-lanemodel based on geometric steering correlations. Alternatively, the vehiclekinematics are considered by a five-body dynamic model. This model iscombined by a semi-empirical tyre model for realistic modelling of the contactforces and torques between the tyre patch and the road. Finally, sensor modelsfor radar, lidar and camera are added to the vehicle model.

To investigate real urban traffic scenarios an advanced driver model isincluded, which uses a pure pursuit path tracking algorithm to follow a giventarget trajectory. To investigate real pedestrian interaction, a real personsbehavior is included by motion capturing technologies. Those heterogeneousenvironments are combined by Co-Simulation to get a real-time connection andfinally the entire testbed.

By applying the Co-simulation environment to a typical inner city trafficscenario, the verification of the system functionality is done. The outcome is asafe and efficient virtual city environment, which enables interactioninvestigations between typical traffic participants and highly automatedvehicles. In summary, the paper shows the high potential of virtual Cosimulationenvironments for progressing automated vehicle functionalities.

Efficient methodologies for the development and validation of highly-automated vehicle systems play an increasingly important role in urban mobility. In addition, the requirements for simulation and verification environments are highly increasing due to a wide range of automation functions to be integrated and validated in the near future.

This paper presents a new approach for developing a virtual test environment for highly automated vehicles based ona high-end virtual reality (VR) engine. The basis is a realistic,geo-referenced city model, which is reactively connected in Co-Simulation to a highly detailed vehicle model or even a Vehicle-in-the-Loop test bed. Through specific sensor model systems like camera radar and lidar, the vehicle can interact with the environment and traffic participants like VRU(Vulnerable Road Users) in real-time. The behaviour of real VRUs in the virtual test field is based on high-end motion-capture technology, which enables the integration of highly realistic avatars in real time.

A proof of concept is shown by application of the virtual urban test field for validating an unmanned battery-electric vehicle. Finally, an outlook for the potential use of advanced VR technologies for agile development processes is given.

Recently, virtual realities and simulations play important roles in the development of automated driving functionalities. By an appropriate abstraction, they help to design, investigate and communicate real traffic scenario complexity. Especially, for edge cases investigations of interactions between vulnerable road users (VRU) and highly automated driving functions, valid virtual models are essential for the quality of results. The aim of this study is to measure, process and integrate real human movement behaviour into a virtual test environment for highly automated vehicle functionalities. The overall system consists of a georeferenced virtual city model and a vehicle dynamics model, including probabilistic sensor descriptions. By motion capture hardware, real humanoid behaviour is applied to a virtual human avatar in the test environment. Through retargeting methods, which enable the independency of avatar and person under test (PuT) dimensions, the virtual avatar diversity is increased. To verify the biomechanical behaviour of the virtual avatars, a qualitative study is performed, which funds on a representative movement sequence. The results confirm the functionality of the used methodology and enable PuT independence control of the virtual avatars in real-time.

The need for efficient and reproducible development processes for sensor and perception systems is growing with their increased use in modern vehicles. Such processes can be achieved by using virtual test environments and virtual sensor models. In the context of this, the present paper documents the development of a sensor model for depth estimation of virtual three-dimensional scenarios. For this purpose, the geometric and algorithmic principles of stereoscopic camera systems are recreated in a virtual form. The model is implemented as a subroutine in the Epic Games Unreal Engine, which is one of the most common Game Engines. Its architecture consists of several independent procedures that enable a local depth estimation, but also a reconstruction of a whole three-dimensional scenery. In addition, a separate programme for calibrating the model is presented. In addition to the basic principles, the architecture and the implementation, this work also documents the evaluation of the model created. It is shown that the model meets specifically defined requirements for real-time capability and the accuracy of the evaluation. Thus, it is suitable for the virtual testing of common algorithms and highly automated driving functions.

The need to find alternative urban mobility solutions for delivery and transporthas led mobility companies to devote enormous resources for researchbasedsolutions to increase vehicle safety. This paper documents a virtual approachto investigate the influences of different load states to the vehicle dynamicof light electric vehicle. A model basing on a three-dimensional multibody system was used, which consists of five bodies. By applying methods of multibody modelling the generalized equations of motion were generated. To include the behavior within the contact point between road and vehicle a simplified tire models was added. The implementation of the equations allowed a first validation of the model via simulations. In a final modeling step the simulation results were interpreted in respect of plausibility. Afterwards,the model was simulated numerically to investigate different load states of the vehicle, by applying constant steering stimuli and variable velocities. In sum,the investigated model approach is useful to identify safety relevant parameters and shows the effects of load states to the vehicle dynamics. Furthermore, it behaves plausibly regarding general vehicle dynamics. These results prove the general usability of the model for the development controllers and estimators in driver assistances systems.

Model based engineering is especially for the development of high performingcontrol systems essentially. By means of suitable simplifications, they help topresent technical relationships and express them mathematically. Thereby,active controllers to influence the system behavior could be developed in anefficient and reliable way.

This paper deals with the design of a holistic simulation environment for an ebikewith a wheel hub motor in the rear wheel and a torque and speed sensor inthe bottom bracket. A model-based approach to development using rapidcontrol prototyping is chosen. The model design is chosen similar to the systemdesign of the control system. The interfaces between the main models are alsothe interfaces of the later controller, which makes it easier to implement thesystem afterwards. The engine dynamics has been tested and adjusted on themodel using a driving cycle. A special focus is on the interpretation of thedrivers inputs by the bottom bracket sensor. At the interface between the sensorand the subordinate engine control system, any desired driving condition canbe set for different types of drivers and driving situations by means of differentcharacteristic curves.

The scenarios investigated are derived from typical simulations needed duringthe development of e-bike drivetrains. They focus on the interaction of thehybrid system consisting of human driver and engine torque. Especially thesynchronization of torques and the reaction to fast increasing stimuli wereinvestigated.

The results show a valid performance of the developed algorithm. The e-motortorque oscillates quickly and the synchronization works fine. Additionally, thealgorithm to smooth the pedaling fluctuations and thereby the torquefluctuations work quite well, whereby a smooth torque is implemented. Nextsteps are the integration of supporting modes and the demand-orientatedcontrol.

Die mechatronische Produktentwicklung verfolgt einen innovativen interdisziplinären Ansatz. Dieser zentralisiert die Funktionen des Endprodukts und bricht dabei effektiv und effizient disziplinabhängige Grenzen auf. Bereits vor der Fertigung der ersten Komponente erfolgt die Durchführung umfänglicher virtueller Untersuchungen sowie eine nachhaltige Optimierung des Ressourcenbedarfs. Im Rahmen des Buches „Mechatronische Produktentwicklung im Kontext der Mikromobilität“ präsentiert sich den Lesern ganzheitlich der Zyklus der mechatronischen Produktentwicklung von der Anforderungsbeschreibung bis hin zum exemplarischen Prototypenentwurf. Die praxisnahe Ausrichtung des Buches versetzt die Leser zielgerichtet in die Lage, im Anschluss mechatronische Produkte im übergeordneten Kontext zu verstehen und somit intelligente Lösungsansätze zu verfolgen. Darüber hinaus kommt es zu einer Erweiterung der Kompetenz im Hinblick auf zeitgemäße Entwurfsverfahren mechatronischer Produkte und der Sinnesschärfung für gesamtheitliches Systemverhalten.

The backscattering models of complex traffic targets play a critical role in the current evolvement of over-the-air (OTA) validation test methods and, consequently, the validation and verification of advanced driver assistance systems (ADAS). Furthermore, stimulating the operational behavior of the automotive radar sensor in an OTA vehicle-in-the-loop (ViL) environment requires radar cross section (RCS) and Doppler signatures related to the movement of vulnerable road users (VRUs), especially pedestrians in traffic scenarios. These RCS and Doppler signatures of human motion can be retrieved by evaluating the range-Doppler behavior of the motions from the human extremities. High-resolution RCS measurements in the radial and angular domains can be applied to micro-Doppler models of human motions and significantly improve them. Therefore, this work presents a measurement technique to estimate the radar reflectivity of different scattering points of human body regions in different moving scenarios. The necessary signal processing steps are discussed in detail for assigning the corresponding RCS values and Doppler signature to the time-varying scattering points from the different extremities of a human body. The measurement results can be used to develop more realistic models of a person in motion in a real traffic scenario which is worthwhile in radar channel wave propagation, target recognition, and classification.

Steer-by-wire systems represent a key technology for highly automated and autonomous driving. In this context, robust steering control is a fundamental precondition for automated vehicle lateral control. However, there is a need for improvement due to degrees of freedom, signal delays, and nonlinear characteristics of the plant which are unconsidered in the design models for the design of current steering controls. To be able to design an extremely robust steering control, suitable optimal models of a steer-by-wire system are required. Therefore, this paper presents an innovative nonlinear detail model of a steer-by-wire system. The detail model represents all characteristics of a real steer-by-wire system. In the context of a dominance analysis of the detail model, all dominant characteristics of a steer-by-wire system, including parameter dependencies, are identified. Through model reduction, a reduced model of the steer-by-wire system is then developed that can be used for a subsequent robust control design. Furthermore, this paper compares the steer-by-wire system with a conventional electromechanical power steering and shows similarities as well as differences.

The concept of autonomous driving has gained increasing relevance, leading to a need for the development of innovative drive concepts for motor vehicles. Therefore, this paper presents a model-based optimal multivariable control for the wheel slip, which allows specifying the wheel slip and thus the tire force individually for each wheel. The plant model consists of a multibody two-track model of a vehicle, a tire model, an air resistance model and a motor model. In addition, the contact forces of the individual wheels are calculated dynamically. The resulting nonlinear model is linearized and used for the design of a linear optimal static state space controller with reference and disturbance feedforward. The contact point velocities at the wheels are defined as the controlled variables, since they are proportional to the wheel slip and thus to the driving forces within the operating range of the controller. In addition, the rates of change of the contact point velocities are also chosen as controlled variables to set damping of the closed-loop system. The four drive torques of the wheels represent the control variables. Therefore, a true multivariable control is developed. In the first step of the analysis, the linearized closed-loop system is investigated regarding stability, robustness and its dynamic behavior. The control system shows a high bandwidth, a well damped dynamic behavior and a large phase margin. In the second step of the analysis, various simulations of realistic experiments, such as an accelerated cornering maneuver or the Fishhook road test, are performed with the nonlinear closed-loop system. The results of these experiments confirm the high robustness and good dynamic behavior of the control system in most cases. Moreover, the results demonstrate how the control considers the dynamic contact forces of the wheels to achieve the requested wheel slip at any time. Lastly, dominant transfer paths are identified based on the gain matrix of the state space controller, showing which input and state variables have significant influence on the control variables. Based on this, single-input single-output controls for the individual wheels can be derived.

The knowledge explosion in all areas is both an opportunity and a risk. In order to shape the effects as positively as possible and create sustainable added value, the amount of information must be processed in a targeted manner and important content must be separated from ignorable content.

In the automotive industry and especially in the development of new components, it is possible to look back on many past projects and thus knowledge. However, the decisive factor is whether this information is available in a suitably processed form or the knowledge is even held by just a single expert. A new and intelligent method is therefore required to analyze existing data appropriately and at the same time prepare it ideally for further applications, such as use within forecast models based on Artificial Intelligence (AI). To achieve this, several steps need to be taken.Firstly, it is possible to perform a suitable segmentation of the component. The aim is to detect areas in a component where features and form elements are found. Other remaining regions are ignored after the inspection by segmentation and voxelization: There is no sustainable valuable knowledge here.

Subsequently, the voxelization of the component takes place, which results in the three-dimensional component or Computer-Aided-Design (CAD) file being mathematically readable and thus a kind of translation takes place. This is done by rasterizing the component based on a previously selected resolution and other upcoming steps.

Finally, the segmented and relevant areas are analyzed accordingly. This can be done according to corresponding previously defined guidelines or, for example, by using a suitably trained AI. The advantage is based in the mathematical readability, which is now given by the voxelization that has taken place.