High speed driving stability of passenger car under crosswind effects

The driving stability of a passenger car at high-speed and under crosswind conditions is affected by aerodynamic characteristics as well as their dynamic characteristics, suspension, and weight distribution. In this study, the total measuring system was thought up to understand the transient vehicle dynamics and aerodynamics with driver’s control inputs all together. The test results were taken from a full-scale wind tunnel test, a crosswind generator test and an on-road test. We investigated major aerodynamic parameters that affect the driving stability of passenger cars under crosswind effects such as overtaking, passing each other, natural crosswind, etc. The reaction rate of high-speed stability will be improved when we minimize the total lift, side force and especially the yawing moment.     

Parallel model based fault detection algorithm for electronic parking brake system

This paper describes a parallel model-based fault detection algorithm for an electronic parking brake (EPB) system, which consists of an electronic control unit with built-in current sensor and braking force sensor. For the EPB system to supply sufficient parking force to a vehicle, the parking force sensor is of utmost importance. If a fault occurs in this sensor, sufficient parking force may not be supplied, thereby seriously threatening the safety of the vehicle. Thus, a fault detection method is required for the parking force sensor of the EPB system to improve the safety of vehicles. For this purpose, a highly reliable fault detection method is needed to detect abnormal fault signals, which cannot be detected by the existing on-line sensor monitoring fault detection methods. This paper proposes a novel parallel model-based fault detection algorithm for the EPB system, which compares the physical sensor data with the mathematical model, the fuzzy model, and the neural network model at the same time. In order to reduce false alarms, the magnitude of thresholds and the operation counts are changed adaptively. When the proposed parallel model-based fault detection algorithm detects severe failures of the force sensor, it warns the driver in advance to prevent accidents due to the failures. The proposed algorithm is verified by hardware-in-theloop simulations in various situations.     

Innovative design optimization strategy for the automotive industry

In order to effectively solve modern automotive design problems including the results of nonlinear FEA and multi-body dynamics, a progressive meta-model based design optimization is presented. To reduce the number of initial sample points, two sampling methods are introduced. Then, for efficient and stable construction of meta-models, three metamodel methods are newly introduced which are numerically based on the singular value decomposition technique. To design a practical system considering manufacturing tolerances and optimizing multiple performances, a robust design optimization, 6-sigma constraints and multi-objective strategies are implemented when solving the approximate optimization problem constructed from the meta-models. Until the convergence criteria are satisfied, the initially developed meta-models are progressively improved by adding only one point that minimizes the approximate Lagrangian in the consecutive optimization iterations. Finally, one validation sample and four automotive applications are solved to show the effectiveness of the proposed approach.     

Automatic parking of vehicles: A review of literatures

Collision accidents often occur during parking or reversing cars. In allusion to this point, this paper conducts a review of literatures on automatic parking. To begin with, a brief introduction of automatic parking including its background and significance is given. Then its commercial application, research status and latest progress are summarized which include visual perception, ultrasonic sensors and radar technology, path planning, control algorithms based on fuzzy theory, neural network, image processing and recognition technology, and digital signal processing technology, etc. On further analysis, some reasonable conclusions are drawn and the future work is suggested.     

Design of automotive headlamp with high-power LEDS

A novel vehicle headlamp low beam system is presented. The optical system is composed of three identical cells to ensure enough illumination. A poly-ellipsoid reflector and a aspherical lens are used to concentrate the rays and a baffle is employed to produce the cut-off line. The optical performance shows a good agreement with the requirement for headlamp in Chinese national standard. The smart constant current driver sustains automotive headlamps?requirements, moreover it has high power efficiency. The performance of three kinds of cooling systems with different structures are simulated and evaluated by FloTHERM software. Then a loop heat pipes combined with fins cooling system is manufactured for cooling the designed LED headlamp. It can greatly improve the cooling efficiency. Even in the 80°C environmental temperature, it can meet the working requirements of LED headlamp low beam system, so it is a kind of feasible cooling scheme.     

Empirical analysis on the effect of design variables of automotive seat lumbar support on the initial sitting discomfort

An empirical analysis was performed in order to examine whether the height of the lumbar support installed on the driver’s seat in a vehicle makes a difference to the initial discomfort when sitting, and among the design variables of the seat lumbar support, which one is the main variable that affects the initial discomfort most when sitting. The subjects who participated in this study were 50^th percentile male in their 30s and 40s from Korea and America, had driving experience, and had not experienced any back pain in the last 12 months. In order to figure out the level of discomfort felt by a driver depending on the movement or the shapes of the lumbar support, subjective sensibility evaluation was conducted and sitting body pressure was measured, and the change of lumbar spine angle was observed using X-ray image. Based on the results, design variables that affect the initial discomfort when sitting on a car seat and the correlation among these design variables were verified through statistical significance testing. The empirical analysis suggested that, among design variables of car seat lumbar support, the degree of prominence is the main variable that affects the initial discomfort when sitting for both Koreans and Americans, while the height and support width have very little to do with it.     

Minimizing vehicle post impact path lateral deviation using optimized braking and steering sequences

This paper investigates the optimal control of a vehicle, after a light impact during a traffic accident. To reduce the risk of secondary events, the control target is set: to minimize the maximum lateral deviation from the initial path. In previous analysis path control was achieved by the active control of individual wheel braking. The present paper examines potential benefits from the additional control of front steering angles. Numerical optimization is used to determine optimal control sequences for both actuator configurations. It is found that steering provides significant control benefits, though not for all post-impact kinematics. For all cases considered, the optimal control operates at the boundary of the control domain of available forces and moments. This domain is expanded when steering is available, and there exists an expanded range of conditions for which coupled control of yaw moments and lateral forces is the most effective control strategy. The sensitivity of vehicle response to the individual actuator controls is studied; it reveals this sensitivity is related to the actuator bandwidth and the lack of any dynamic cost in the longitudinal direction. This motivates a further analysis which includes longitudinal and lateral dynamics in the cost function. This is broadly related to real-world crash risks. Further, different versions of such cost functions are compared as a basis for implementation in a closed-loop controller.     

Development of active front wheel steering control system keeping stable region in driving phase diagram

This paper presents a new active steering control system based on driving phase diagram (β _fr ?δ _f diagram). In order to make state variables to follow those of nominal vehicle model that was developed under no consideration of disturbance, Quadratic Programming Problem (QPP) is formulated, where time varying objective function minimizes the differences between nominal and actual parameters. The steering characteristic in active steering control system changes when the vehicle faces disturbance such as crosswind and flat tire, and driver tries to counteract it after recognizing the change. The proposed method defines a stability region on β _fr ?δ _f diagram. In order to make β _fr and δ _f remain in the stability region, a new model predictive controller is proposed. While conventional controllers are restrictive to satisfy the β _fr ?δ _f diagram based stability condition, the proposed controller ensures solution space and also plays a direct role to minimize the evaluation function in the constrained optimal control problem.     

New tyre-road contact model for applications at low speed

Most of the tyre models have been developed for high speed, combined forces, etc., however, in certain tests it is necessary to know tyre behaviour at very low speed in order to evaluate different systems. So, during vehicle inspection and maintenance of the steering and brake system, by means of sideslip tester and roller brake tester respectively, the forces transmitted by the tyres are measured; all of these inspections are carried out at low speeds. Furthermore, usually, automobile vehicles run at low speeds during an important part of their operating life (less than 60 km/h), mainly during urban traffic, and in steady state conditions. Therefore, it is particularly interesting to develop an accurate model of the contact patch tyrepavement for low speeds without the complexity of models that cover a wide speed range but provide less precision at very low speeds. The dynamometer plate has proved to be an appropriate test equipment to characterise the tyre-pavement contact at low speed and the steering geometry and wheel alignment. It has the feature of being able to carry out tests with the tyre installed in the vehicle as in completely real conditions. The main aim of this research is to set up a contact model between tyre and pavement at very low speed based on the measurement of longitudinal and lateral forces. A test methodology that allows carrying out the experimental tests in a systematic and controlled way with the dynamometer plate has also been developed. From this model it will be possible to estimate the forces that tyres are capable of transmitting in different situations to act in the parameters which affect these forces and maximize them.     

The validation of a microscopic simulation model: a methodological case study

The use of microscopic simulation models to assess the likely effects of new traffic management applications and changes in vehicle technology is becoming increasingly popular. However the validity of the models is a topic of increasing concern, as the quality of the presentation often exceeds the models ability to predict what is likely to happen.Traditionally, model validity has been ascertained through comparing outputs aggregated at a macroscopic level such as speed flow and lane use, against real data. Little microscopic comparison is generally possible and, where this is done there is often no separation of the calibration and validation process. This paper demonstrates how microscopic validation may be undertaken when suitable data is available, in this case time series data collected by an instrumented vehicle, and its use in the validation of the car following performance of a fuzzy logic based car following model. Good agreement has been attained between the simulated model and observed data, primarily using a root mean square error indicator. Lastly, a brief comparison of the new model with the performance of a number of existing formulations has also been undertaken.