Search Results

This paper describes a novel approach for modeling an automotive HVAC unit. The model consists of black-box models trained with experimental data from a self-developed measurement setup. It is capable of predicting the temperature and mass flow of the air entering the vehicle cabin at the various air vents. A combination of temperature and velocity sensors is the basis of the measurement setup. A measurement fault analysis is conducted to validate the accuracy of the measurement system. As the data collection is done under fluctuating ambient conditions, a review of the impact of various ambient conditions on the HVAC unit is performed. Correction models that account for the different ambient conditions incorporate these results. Numerous types of black-box models are compared to identify the best-suited type for this approach. Moreover, the accuracy of the model is validated using test drive data.

Highly automated driving (HAD) is under rapid development and will be available for customers within the next years. However the evidence that HAD is at least as safe as human driving has still not been produced. The challenge is to drive hundreds of millions of test kilometers without incidents to show that statistically HAD is significantly safer. One approach is to let a HAD function run in parallel with human drivers in customer cars to utilize a fraction of the billions of kilometers driven every year. To guarantee safety, the function under test (FUT) has access to sensors but its output is not executed, which results in an open loop problem. To overcome this shortcoming, the proposed method consists of four steps to close the loop for the FUT. First, sensor data from real driving scenarios is fused in a world model and enhanced by incorporating future time steps into original measurements.

The aim of the project is to reliably identify the set of constructive features responsible for the highest noise levels in the interior of motor vehicles. A simulation environment based on artificial intelligence techniques such as neural networks and genetic algorithms has been implemented. We used a system identification approach in order to approximate the functional relationship between the target noise series and the sets of constructive parameters corresponding to the cars. The noise levels were measured with a microphone positioned on the driver''s chair, and corresponded to a variation of the engine rotation of 600-900 rot/min. The database includes 45 different cars, each described by vectors of 67 constructive features.

The Advanced Lighting Simulation (ALS) is a development tool for systematically investigating and optimizing the Adaptive Light Control (ALC) system to provide the driver with improved headlamps and light distributions. ALS is based on advanced CA-techniques and modern validation facilities. To improve night time traffic safety the BMW lighting system ALC has been developed and optimized with the help of ALS. ALC improves the headlamp illumination by means of continuous adaptation of the headlamps according to the current driving situation and current environment. BMW has already implemented ALC prototypes in real vehicles to demonstrate the advantages on the real road.

A local Gaussian process regression approach is presented, which allows to model nonlinearities of internal combustion engines more accurate than global Gaussian process regression. By building smaller models, the prediction of local system behavior improves significantly. In order to predict a value, the algorithm chooses the nearest training points. The number of chosen training points depends on the intensity of estimated nonlinearity. After determining the training points, a model is built, the prediction performed and the model discarded. The approach is demonstrated with a benchmark system and air charge test bed measurements. The measurements are taken from a turbocharged SI gasoline engine with both variable inlet valve lift and variable inlet and exhaust valve opening angle. The results show how local Gaussian process regression outmatches global Gaussian process regression concerning model quality and nonlinearities in particular.

Air charge calibration of turbocharged SI gasoline engines with both variable inlet valve lift and variable inlet and exhaust valve opening angle has to be very accurate and needs a high number of measurements. In particular, the modeling of the transition area from unthrottled, inlet valve controlled resp. throttled mode to turbocharged mode, suffers from small number of measurements (e.g. when applying Design of Experiments (DoE)). This is due to the strong impact of residual gas respectively scavenging dominating locally in this area. In this article, a virtual residual gas sensor in order to enable black-box-modeling of the air charge is presented. The sensor is a multilayer perceptron artificial neural network. Amongst others, the physically calculated air mass is used as training data for the artificial neural network.

Nowadays, a continually growing system complexity due to the development of an increasing number of vehicle concepts in a steadily decreasing development time forces the engineering departments in the automotive industry to a deepened system understanding. The virtual design and validation of individual components from subsystems up to full vehicles becomes an even more significant role. As an answer to the challenge of reducing complete hardware prototypes, the virtual competence in NVH, among other methods, has been improved significantly in the last years. At first, the virtual design and validation of objectified phenomena in analogy to hardware tests via standardized test rigs, e.g. four poster test rig, have been conceived and validated with the so called MBS (Multi Body Systems).

Handling characteristics, ride comfort and active safety are customer relevant attributes of modern premium vehicles. Electronic control units offer new possibilities to optimize vehicle performance with respect to these goals. The integration of multiple control systems, each with its own focus, leads to a high complexity. BMW and ITK Engineering have created a tool to tackle this challenge. A simulation environment to cover all development stages has been developed. Various levels of complexity are addressed by a scalable simulation model and functionality, which grows step-by-step with increasing requirements. The simulation environment ensures the coherence of the vehicle data and simulation method for development of the electronic systems. The article describes both the process of the electronic control unit (ECU) development and positive impact of an integrated tool on the entire vehicle development process.

When investigating the posture comfort in vehicles two important influencing factors can be distinguished: In order to evaluate these influences a combined laboratory-field-experiment was carried out. A real car was equipped with cameras to record the body posture and the joint angles. The static forces exerted by the driver on his contact points were recorded in a corresponding mock-up. The forces to maintain the body posture were calculated. The following results were found:

In this paper a method for the identification of most significant vehicle parameters influencing the behavior of a lateral control system of autonomous car is presented. Requirements for the design stage of the controller need to consider many uncertainties in the plant. While most vehicle properties can be compensated by an appropriate tuning of the control parameters, other vehicle properties can change significantly during usage. The control system is evaluated based on performance measures. Analyzed parameters comprise functional tire characteristics, mass of the vehicle and position of its center of gravity. Since the parameters are correlated, but Sobol’ sensitivity analysis assumes decorrelated inputs, random variation yields no reasonable results. Furthermore, the variation of each parameter or set of parameters is not applicable since the numbers of required simulations is increased significantly according to input dimension.