10th International Munich Chassis Symposium 2019

∎

∎

∎

∎

Current and future vehicle development alike will be shaped by two megatrends. Firstly, there is the electrification of the powertrain, which is intended to further reduce and shift traffic-related CO2 emissions and immissions. Concepts in this field range from purely electric powertrains right through to complex hybrid topologies.

∎

∎

The process of urbanization is ongoing with emphasis in Asia [1]. The number of megacities with more than ten million inhabitants has been growing rapidly for years, particularly in Asia. The result of this process is a specific need for mobility, making special vehicle concepts necessary.

The electrification process which has been started in the automotive market opens a lot of opportunities to the engineers to think again to the best solution they can design in order to have the best result in terms of functionality, performance and costs. As the constraints in an battery electric vehicle change, also the answers to the questions every engineer has during the development can be different in comparison to the standard solutions which are now widespread in the market. This revolution doesn’t affect only the powertrain development, the controls and the human-machine interface, but could affect all the fields present in a car development.

Wet roads dominate the traffic accident statistics, well above all other weather-related factors. Apart from speed-induced hydroplaning situations, drivers tend to underestimate the effect of a wet pavement on braking distance or overall vehicle stability.With the Porsche Wet Mode, the new generation of the 911 introduces a driver assistance system, designed to flag the road condition up to the driver, along with offering a stability setup for wet pavements.The detection of the road condition relies on acoustic sensors attached to the wheel arch liners of the front axle. When driving over a water film, the sound of the road noise and the impact of the spray dispersed by the tires triggers the acoustic sensors and urges preconditioning of the chassis control systems. When the situation persists or the spray intensity increases, the instrument cluster suggests to the driver to activate a special drive mode, which facilitates driving on wet pavements.

Simulationen und Tests liefern oftmals wichtige Erkenntnisse über Eigenschaften eines Produkts. Welche Eigenschaften überprüft werden müssen, hängt davon ab, welche Eigenschaften das Produkt überhaupt aufweisen muss. Aus rechtlicher Sicht ist der anzulegende Maßstab hier zunächst der vertraglichen Vereinbarung mit dem Käufer des Produktes zu entnehmen. Dies reicht aber nicht aus, da es oftmals weitere Erwartungen an ein Produkt bestehen: Erwartungen, die üblicherweise an das Produkt gestellt werden bzw. Erwartungen, die die Sicherheit des Produktes betreffen.

Vehicles are subject to a variety of requirements, of which the technical objectives are only a subfield. In addition, the different sub-areas increasingly interact with each other, whereby the complexity of the development process of a vehicle continuously rises. In particular, components with a high degree of interaction must therefore be considered overall with regard to all influences.

The aim of real-time models is, on the one hand, to improve driving dynamics and ride performance and, on the other hand, to be able to be used in driving simulators. In order to achieve these goals, not only a short computation time is necessary, but also the model accuracy is very important. For this reason, this paper investigates the simulation results of a developed real-time model and compares them with results from physical tests and a multibody simulation model. The influence of the modeling level on the simulation results is discussed in the context of the investigations carried out. For quasistatic analysis the Kinematic & Compliance (K&C) test rig and for dynamic tests a dynamic suspension test rig is used.

In the current paper a model-based approach for vehicle parameter identification will be proposed. In order to increase the quality of the simulated vehicle behaviour (responses like lateral acceleration, yaw rate and especially the steering rack force) the best fitting combination of the vehicle mass inertias, the metacentric height as well as some selected tire parameters (like cornering stiffness, pneumatic trail, etc. [4, 9, 10]) needs to be identified.

Modern chassis control systems, advanced driver assistance systems (ADAS) and automated driving systems that demand a precise vehicle localization or a reasonable trajectory planning desire a highly accurate and reliable vehicle state estimation. However, the traditional methods such as Kalman and RLS filter, which based on the vehicle dynamic model, mainly rely on the differential equations that approximate the vehicle behaviour in reality [1, 27, 31]. The vehicle dynamics is such a nonlinear and multidimensional system with numerous parameters, which makes it very difficult to adapt the parameters in different situations and figure out appropriate model equations.

The active suspension system has always been a topic of interest because of its ability to influence the ride quality by exerting independent forces on the suspension by the usage of separate actuators. Various strategies have been proposed over the years in order to estimate the appropriate control action. These strategies are typically feedback oriented and dependent on many factors like the control objective, the frequency of the excitation and system non-linearities that result in the formulation of a complex problem. A simulation-based study using a comprehensive quarter car model with an active suspension is beneficial to summarize the character of each of these approaches by comparing attributes like formulation, performance, robustness, tunability and requirements for implementation on real physical systems.Since active suspensions use an on-board computer and sensor measurements to determine the control action, the enactment of a strategy is limited by the computational requirements and available measurements. This study aims to foresee such challenges when implementing modern control strategies like H∞ control or preview based approaches like Model Predictive Control (MPC) and Deep Learning methods. The latter will focus on both Supervised Learning (SL) and Reinforcement Learning (RL) approaches. These methods are firstly developed in a virtual environment and subsequently implemented on a physical quarter car setup excited by a servo-hydraulic actuator. Finally, a comparison of the performances of the different control approaches is presented.

Deep Learning based behavior reflex methods found their way into modern vehicles. To model the human driving behavior it is not sufficient to rely solely on individual, noncontiguous camera frames without taking vehicle signals or road specific features into account. In this work four temporal fusion methods are evaluated based on three different Deep Learning models. The proposed spatio-temporal Mixed Fusion model extends the present end-to-end models and consist of multiple levels of fusions. The raw image data from a single front facing camera is mixed with recorded vehicle data and a map based predicted road bank angle gradient vector. The model accesses multiple time axes: temporal features of multiple image frames are extracted through a combination of Convolution and LSTM layers while it can also make assumptions about the future road condition with the use of upcoming Ground Truth road bank angle changes. Experiments are performed on a recorded data set of real world drivings. Results show, that this approach leads to an accurate imitation of the human driver with an inference capability of more than 60 FPS.

□

The use of composite materials e.g. carbon fibre reinforced polymers in suspension systems can contribute to lighter and therefore more energy-efficient passenger cars and commercial vehicles. Besides ensuring the technological feasibility, the implementation of the corresponding large-scale production is essential to compete with classical metal structures. Within the chassis, both lateral and longitudinal leaf springs made from composite materials have been found to meet these requirements. Single- and multi-axis struts, produced in a spatial winding process, are a further promising example of load-path optimized chassis components showing high potential for automotive series applications.In the present paper, both composite leaf spring and strut applications are presented with a focus on the holistic development approach besides their respective technical features.

In strut front suspensions, a sickle-shaped triangular control arm is often used as a lower suspension arm for small and mid-sized vehicles. With the growing importance of electric vehicles, the demands for driving comfort and acoustics are increasing. These requirements force a reduction of the rubber bushing stiffness, which cannot be properly realized with the current bushing arrangement of the sickle-shaped triangular control arm. For a solution to the goal conflicts it is necessary to clearly separate the lateral stiffness and longitudinal elasticity functions in the lower suspension arm. Previous implementations use the resolution of the triangular control arm in two separate suspension arms. To reduce costs, a metal-elastic triangular control arm is developed which combines the functions mentioned above in one component. The concept consists of a compression strut region, which is responsible for the absorption of longitudinal and proportional lateral forces and a transverse link area, which transmits lateral forces along its alignment, but allows a displacement of the front wheel suspension in the vehicle longitudinal direction by its metal elasticity.In this paper, based on the state of the art, the procedure of the design of the metalelastic triangular control arm is explained. The characteristics of the front wheel suspension with the sickle-shaped and the metal-elastic triangular control arm are shown in a functional data comparison.

Nowadays, a huge amount of research work relates to autonomous driving. Nevertheless, the degree of automation in current series-production cars is limited to the 2nd and to the 3rd level according to SAE J3016 [1]. The electrical powered steering system is an essential component of these cars because it provides controllability for the driver and assist for comfort and safety reasons. In contrast, the relating motor torque is able to harm other cars or even pedestrians in case of malfunction.

Advanced Driver Assistance Systems (ADAS) and Highly Automated Driving Systems (HAD) are among the most important megatrends in the automotive development. Accompanying this one big question arises: do all assisted and automated driving cars drive the same or will vehicle manufacturers be able to differentiate themselves with DNA of their own? And - especially for a sportscar-manufacturer like Porsche – how can ADAS and HAD impart typical attributes like driving fun and sportiness, Figure 1? In order to achieve this, clear driving characteristic goals (from a customer’s point of view) must be defined and the system requirements for ADAS and HAD (including all components like sensors, ECU’s and actors) shall be derived from this. However, what are driving and brand characteristics in the context of assisted and automated driving? And how can those be realized in the development? Porsche has addressed this question together with the University of Applied Sciences Kempten and MdynamiX.

More and more sophisticated assisted/autonomous vehicles are becoming available in the market. Automation levels 2 and 3 have been given for settled just a couple of years ago, and the path to fully autonomous car seemed to have no obstacles. In reality, OEMs started recently realizing that the impact of semi and/or fully robotized cars on drivers as well as on passengers is all but predictable. VI-grade has more than ten years experience with developing turn-key solution driving simulators, and has been working for more than five years on a research project to collect meaningful bio-signals from the driver during simulator sessions, in collaboration with the BACPIC of the Catholic University of Sacred Heart and the DPIA of the University of Udine. Recently, a collaboration with the Human Inspired Technology Research Center of University of Padova allowed the extension of the assessment at physio-emotional level.

Full autonomous vehicles for the general public are getting closer every year. Among all the challenges to overcome, one of them is the acceptance of this technology which translates to make the passengers enjoy being driven. To achieve this objective, automated vehicles will have to focus on performance attributes such as comfort, stability or efficiency and vehicles dynamics development will take care of it.At Idiada, research about this topic is being carried out and strategies about motion planning and control (path follower) will be proposed in this paper. The development is based on optimal control method like Model Predictive Control (MPC).Among all the possibilities to face the problem, MPC was chosen for several reasons. MPC allows setting constraints in our control inputs like maximum steering wheel angle or vehicle states like accelerations. However, the main reason to use MPC is its way of planning in advance control actions that behaves very similar to how a human driver would do. This feature is key in our understanding to make autonomous vehicles be accepted by all passengers.Also, our contribution yields in finding the correct vehicle dynamics metrics to design and adjust all the cost functions.Finally, thanks to our new acquisition, the DIM 250 VI-Grade Simulator which is able to reproduce up to 2.5G accelerations, all our development will be evaluated in a fast and secure testing environment.

Legal and regulatory pressures in Europe and China to reduce emissions and fuel consumption are increasing the demand for hybrid and electric vehicles. This growing demand is the main driver for technological improvements in electric powertrains and batteries. Nearly every car manufacturer is developing hybrid or electric vehicles to fulfil the fleet consumption targets in order to avoid penalties.

The introduction of highly automated driving features will lead to major changes in the vehicle’s E/E architecture: among other things, a reliable infrastructure will be indispensable. Currently, the driver is responsible for the vehicle at any time and serves as a backup for possible functional failures. With the introduction of highly automated driving functions, this will no longer be the case in the future, as the driver can have other activities in the car like reading books, working or even sleeping. Psychologists have determined how long a driver needs to finish his job and regain control of the vehicle: about 15 seconds are needed for this change [1], [2], [3].

Brake system dynamics are governed by highly nonlinear effects when required to propagate high hydraulic pressure within a short amount of time to the brake calipers. Due to substantially more demanding conditions and requirements in a Formula 1 car’s brake system do not only these effects have a bigger impact, but make it also more difficult to robustly control the brake caliper pressure when brake-by-wire is adopted. The control system presented in this paper originates from Alfa Romeo Sauber F1 Team’s desire (today Alfa Romeo Racing) for an alternative algorithm with enhanced robustness and similar performance when compared to the existing proportional- integral based one. To this purpose, an innovative nonlinear controller based on back-stepping methods and adapted robust control has been developed. Its ability to cope for model uncertainties and unknown disturbances, while guaranteeing stability over the full nonlinear operational range and asymptotic error convergence, are the focus of this work alongside the illustration of the controller’s working principles and the small step to reach nonlinear vehicle dynamic control.

Providing adequate ride comfort at premium class passenger cars is essential to meet the costumers’ expectations. A key feature of “good ride comfort” is the capability of the vehicle’s suspension system to isolate the vehicle body – and thus the passengers – from road imperfections and undulations, which otherwise cause unpleasant oscillations and accelerations. The respective low-frequency range, often referred to as primary ride, [1], is of main interest within this paper. Passive suspension systems have evolved to a high Ievel of maturity, and due to the particular subjective nature of the human perception of ride comfort, this achievement is largely based on the experience of generations of test engineers and their efforts on vehicle testing on proving grounds and/or public roads. No further substantial improvements are tobe expected for passive spring and damper, [2, 3].

Testing and evaluating of dampers, top mounts and its assemblies, can be challenging when it comes to HiL-applications and characterization in the small signal range with real-time capability. Especially the real-time environment has to cope with signal input, signal recording, vehicle model simulation and the control of actuators at the same time to meet the specification of hard real-time capability. Another challenge is the low velocity identification of small damper forces which are important for adequate friction modelling. In this paper a successfully implemented concept of a chassis HiL test bench will be presented. This chassis HiL test bench is designed especially for ride comfort tests in the small signal range. Although there are some limitations due to actuator dynamics, it is possible to carry out characterization and HiL tests with small displacement and velocity amplitudes. A fast actuator control is possible by implementing a fiber optic based communication system which guarantees 125 μs latency. Furthermore, test scenarios based on advanced characterization and HiL tests are presented. These tests scenarios were developed for tests and ride comfort investigations in the small signal range. Using the example of a rear axle twin-tube damper assembled with a top mount, these test scenarios show very good results. With this chassis HiL test bench and the tests scenarios more accurately test results for the product development process of passive, semi-active and active dampers, top mounts and its assemblies are possible.

Vehicle dynamics is a fundamental part of vehicle performance. It combines functional requirements (i.e. road safety) with emotional content (“fun to drive”, “comfort”) [1, 2]. The balance of Ride & Handling is what often characterizes the car manufacturer (OEM) driving DNA. The level reached nowadays by most of the OEMs on Ride & Handling is very high. This has been possible combining experience of years of development with virtual vehicle simulation. Vehicle simulation allows finding the best solutions yet in the pre-development phase, in purely virtual stage. Simulation tools can reduce significantly development costs, fundamental from OEM prospective, and they can still be useful to support the physical development of the vehicle (i.e. tuning).One of the most important chassis components, contributing to define the character of the vehicle, is the damper [3]. Despite of its importance, most of the OEMs are mainly relying on supplier’s “know-how” about design and on experienced drivers about tuning. Usually 1D lookup tables Force vs. Velocity, generated from tests like the standard VDA, are normally used by OEMs to describe the damper behavior. This approach is not representative of the damper’s construction. This modelling is not able to describe how the damper is actually developing the force, mainly dependent on valve topology and tuning strategy. Different dampers display the same Force vs. Velocity curve but they can give different feeling to the driver. Consequently, the capability to represent the full damper behavior, in testing and numerical simulation, is fundamental.To do that, an advanced CAE damper model (starting from Duym research [4]) and a new testing protocol have been developed in collaboration between Hyundai Motor Group and Politecnico di Torino. The virtual model has been developed in Matlab/Simulink® to be easily integrated in co-simulation with the CAE process used in HMETC (e.g. Driving Simulator).It represents the damper behavior by the physical properties of its components (such as rod, valves components, oil properties [5], friction [6], etc.). Most of the parameters are sourced directly from damper drawings, Bill of Material or by measurements. The model has been tuned and verified against the output of the testing protocol, showing a good level of correlation up to 30 Hz. This is very important to correlate the vehicle ride up to high frequencies. Because of that, it is possible to use the driving simulator not only for Handling but also for Ride vehicle evaluation.The testing protocol has been defined in order to have more details on how the damper is developing the force on selected range of component speeds & frequencies, under quasi-static and dynamic conditions. Moreover, a new way to analyze results in frequency domain has been proposed, to better understand, describe and correlate the damper performance to whole vehicle behavior.

High available steering systems for autonomous driving enable the development of Steer-by-Wire systems. Due to the missing mechanical linkage the steering wheel torque which depends on several inputs with different influences needs to be artificially generated by a Force-Feedback-Actuator.The present publication describes how machine learning methods and especially artificial neural networks can be used to provide a steering feel for a Steer-by-Wire system. Therefore training data consisting of synthetic driving maneuvers is recorded to train a feedforward neural network. Measurement signals of the driver input and the vehicle reaction were selected to be used as inputs for estimating the steering wheel torque as the output of the model.Networks of different sizes are trained and evaluated on the basis of their training and test error to examine how complex the model must be to calculate the output sufficiently. To extract more information from the training data sliding window features are used in addition to the current signal values.A trained network has been integrated into the software of a Steer-by-Wire system in a prototype vehicle to provide the steering wheel torque to the Force-Feedback-Actuator. In this vehicle the steering feel generated by the model could be subjectively evaluated on a test site.

For the control of electric steering systems, there is still a need for a robust implementation due to unconsidered degrees of freedom, nonlinear spring stiffnesses and gear ratios. Therefore, this article describes the modeling, control design and system analysis of an electric power steering system. It is focused on the degree of robustness of the possible controls. Two LQG-controllers are designed which are able to perform active vibration damping and disturbance compensation. This allows a high control bandwidth. Both resulting control systems are robust against unconsidered degrees of freedom, nonlinear system behavior and variations of plant parameters. Thus, the presented controllers fulfill the requirements of a modern steering system and allow the adaptation of the steering feel to the current driving situation.

Vehicle manufacturers try to enhance driver to steering dynamics as a key product power element for their customers. In parallel, on the long path to fully autonomous driving, these dynamics will become more complex in the interaction with driving aid systems and as driver’s engagement gets more intermittent.

ADAS and AD are the two major topics for the current and next generations of vehicles. In this context, it is necessary to develop technical solutions to enable larger-size vehicles to follow the trend with an existing 12 V vehicle power supply. EPS can only replace HPS (Fig. 1) as long as this does not degrade performance during any steering maneuvers.

□

By 2050, 70 percent of the world’s population will live in cities as a result of ongoing urbanization. This leads to various challenges such as increasing traffic within cities and less available space. Therefore, multidirectional vehicles with high maneuverability are an advantageous vehicle type for urban areas.

The straight running of the system composed by a car plus driver is studied. Straight running is an important case study for analysing stability. Despite the lateral slip angles of the tyres are small, the system is highly non linear, due essentially to the driver action. Following the simple model of McRuer, later developed by Mistschke and revised by many other authors, we have developed a mathematical model of a car plus driver. The dynamic behaviour of the mathematical model has shown the presence of limit cycles generated by so called Hopf-bifurcations. The mathematical model predicts that, despite the understeering vehicle is globally stable, the driver can make the whole system (car plus driver) unstable. This occurs in case an external disturbance is sufficiently strong. If the external disturbance is small, the understeering vehicle plus driver remains stable. There is a speed above which the understeering car plus driver is unstable, usually such a speed is much greater than the maximum speed of the car on high grip surface. The statements introduced above have been validated by employing the driving simulator of Danisi Engineering, Nichelino, Italy. We experimentally saw that limit cycles do exist and that the driver can make the understeering vehicle model of the simulator quite unstable. We were able to validate the mathematical model by including two humans in the driving loop. One driver was a professional driver, the other one was a novice. The same non linear behaviours were highlighted for the two drivers, however, the amplitudes of the limit cycles and the ability of controlling the car were higher for the professional driver. A question arises whether an electronic power steering (EPS) may reduce or cancel instability. The answer is that there are a number of possible solutions for ESP to counteract the effect of unstable limit cycle.

In the context of automated driving a variety of different technologies will hit the market. Increasing levels of automated driving will have an influence on functions and styling of future steering wheels. Based on a general view on possible automation levels and boundary conditions, three different automated vehicle types have been identified. Those needs and requirements for steering wheel systems will be discussed in this paper. These range from increased sensing functions for driver state, additional input and communication technologies to new steering wheel shapes and foldable steering wheels.

There are several categories defined by SAE automation levels on the way to autonomous driving [1]. Nowadays, automated functions associated with the 2nd and 3rd level exist in cars, e.g. the lane keeping assist and the traffic jam pilot system. This paper looks forward and focusses on the degree of high automation, respectively the 4th level.

□

∎

The automotive industry is in a radical change because of business fields. Topics like emission free mobility as well as highly automated driving are the new key topics of the industry. To be robust for the future the industry already optimizes its budgets and adapts its development processes to reduce the time to market.

Plasma electrolytic oxidation (PEO) is a relatively new electrochemical surface treatment process for the generation of oxide coatings on metals such as Aluminum or Magnesium. Most of the papers, referring to the PEO process, were published in the eighties of the previous century [2, 3]. Nowadays, the process is used for a wide range of applications in various industrial fields [4]. In plasma electrolytic oxidation higher voltages are applied than in the anodizing process.

With the improvements made on exhaust particle emissions, non-exhaust and especially brake particles are pulled out from the dark. Tallano Technologie has developed a brake particles collection system, named TAMIC. TAMIC was designed to trap at least 80% of brake particles directly at the pad-disc interface without altering braking efficiency.

Cordenka rayon is a high tenacity man-made fiber based on cellulose extracted from wood. Its key property is a high Young’s modulus that is performed constantly in a temperature range between -40 and +120 °C. This feature as well as rayon’s low thermal shrinkage, its good resistance to chemicals and its sustainable raw material origin have made rayon popular as reinforcing material for high performance automotive tires and brake hoses.

Vehicles with electrical propulsion and energy recovery need specific brake systems able to guarantee the correct integration of electrical energy recovery and high braking capability. The use of advanced controls with brake by wire systems allows to integrate such functionalities leaving the correct feedback to brake pedal and the possibility to enhance vehicle dynamics. We present a study done with simulation and simulators with human in the loop to show the potentiality of specific controls to enhance vehicle dynamics and their integration with electric motor controls. In detail, we will show how a physical brake-by-wire unit can be installed on a simulator environment obtaining an enhancement in the driving realism and enabling the virtual development of advanced vehicle functions. The same testing architecture can be used also for fault-injection activities and precise pedal feeling calibration. The case-study test bench has been installed on a static driving simulator equipped with a concurrent real-time machine, a vigrade car real time model, the complete vehicle network and the full steering system.

In 1996, Daimler AG was the first manufacturer who introduced an electronic brake system (EBS) for commercial vehicles above 7.5 t [1]. The brake system includes a function which controls the slip on different axles and therefore replaces the load sensing valve [2], [3]. In comparison to pneumatic brake systems, the EBS has further advantages such as a shorter response time, load independent and inertial referenced deceleration requests and brake wear harmonization.

Starting in 1997, all fuel consumption and emission values in Europe were based on the so-called “New European Driving Cycle (NEDC)”. Although this cycle did not represent the typical vehicle-driver related fuel consumption values, it was very suitable for comparing cars and engines.

The basis for driving safety and driving comfort in light-alloy wheels for cars are the unspring masses, whereby it is crucial that the weight of the wheels is as low as possible. Due to the inertia and the rotational moment one is endeavored to use light wheels. For this reason, on the one hand, attempts are being made to structurally implement lightweight construction on wheels.

∎

Stopping distance measurements are affected by a large quantity of noise factors that contribute to make them not comparable when measured in particular conditions.

The choice of the optimal tire setup for a given vehicle is not a trivial task. Nowadays car manufacturers often collaborate with the tire manufacturers during the development phase of new vehicles in order to improve the quality of their final products. At the beginning of a new development project, vehicle engineers define the so called vehicle key performance indicators (KPIs) and respective targets that the vehicle has to achieve.

The tire as the only contact between vehicle and road surface has a significant influence on vehicle safety, ride comfort, driving dynamics and fuel economy. A tire can reach the optimum performance, only when an appropriate tire pressure has been adjusted. Due to the importance of tire pressure, the USA has published a legal regulation (FMVSS 138) to require that tire pressure monitoring systems have to be installed in light vehicles since 2007.

Due to the high mass and required volume for a high capacity battery, Sports Utility Vehicles (SUVs) with their large dimensions and high dead load of structure currently appear to a suitable vehicle concept for electrification. Mounting the battery under the bottom of chassis leads to larger moments of inertia and lower center of gravity, whose influence on dynamic rollover behavior has not been investigated sufficiently. In addition, the influence of tire characteristics on nonlinear driving dynamic behavior i.e. rollover performance has been studied only partially.

Non-exhaust emissions from on-road vehicles are gaining increasingly public attention. Besides the disk brake also the tires represents a significant source of abrasion particles of road traffic. The tyre tread surface wears out due to frictional contact with the road surface. To this effect, a new measurement facility, unique in its kind, was designed and developed at the Technische Unversität Ilmenau to assess tyre particulate emissions.