Search Results

As the mobility being developed becomes more complex and numerous, it is becoming difficult and inefficient to apply current vehicle-test-based development. To overcome this, research on combining test and simulation models has been actively conducted to perform objective and subjective evaluations more accurately and efficiently in the advance stage without a vehicle over the years. At first, test models for various systems such as tire, suspension and body were made compatible with simulation models by using various methodologies such as blocked forces, FBS decoupling, and Virtual Point Transformation (VPT). The second step was to objectively estimate road noise by using FBS coupling with system models and to deeply analyze transfer paths and system’s sensitivity. The results were verified by comparing with what was measured and analyzed on vehicle.

During icing wind tunnel experiments, the calibration process of the spray nozzle and aerothermal systems introduces experimental uncertainty that can potentially compromise the reliability of the test results. Therefore, performing sensitivity analysis (SA) or uncertainty quantification (UQ) studies is not only essential to determine the influence of uncertainties on the ice shape and aerodynamic performance but also crucial to identify the most significant icing parameter uncertainty. However, given the wide range of icing envelopes, it is not practical to conduct SA and UQ by experimental method because a lot of evaluations are required for SA and UQ study. In this study, we addressed these challenges by using a deep learning-based reduced-order modeling technique.

Diagonalized alternating-direction implicit (DADI) method is implemented in the Eulerian hyperbolic droplet solver, ICEPAC, for efficient high-order accurate analysis of aircraft icing. Detailed techniques for implementing the DADI method considering hyperbolicity characteristics are discussed. For the Eulerian droplet equation system to be strictly hyperbolic, additional source terms regarding artificial droplet pressure are included. Validations of the present implicit solver are conducted using two- and three-dimensional steady benchmark tests: NACA0012 airfoil, NACA23012 airfoil, and a swept wing. Also, the oscillating airfoil SC2110 case was analyzed to verify the robustness and efficiency of the proposed solver. In addition, the computational cost of the current implicit solver is considerably lower than that of the explicit multi-stage solver.

This paper presents a stability monitoring algorithm with a combined slip tire model for maximized cornering speed of high-speed autonomous driving. It is crucial to utilize the maximum tire force with maintaining a grip driving condition in cornering situations. The model-free cruise controller has been designed to track the desired acceleration. The lateral motion has been regulated by the sliding mode controller formulated with the center of percussion. The controllers are suitable for minimizing the behavior errors. However, the high-level algorithm is necessary to check whether the intended motion is inside of the limit boundaries. In extreme diving conditions, the maximum tire force is limited by physical constraints. A combined slip tire model has been applied to monitor vehicle stability. In previous studies, vehicle stability was evaluated only by vehicle acceleration.

This paper presents a methodology of trajectory planning for the surrounding-aware lane change maneuver of autonomous vehicles based on a data-driven method. The lateral motion is planned by sampling candidate patterns which are defined based on quintic polynomial functions over time. Based on the cost evaluation among the sampled candidates, the optimal lateral motion pattern is selected as a reference and tracked by the controller. The longitudinal motion is planned and controlled using Model Predictive Control (MPC) which is an optimal control method designed considering the surrounding traffic information. To realize the lane change motion similar to the human driving behavior in the surrounding traffic situation, the human driving pattern is modeled in the form of motion parameters and considered in planning the lateral and longitudinal motion.

This paper describes a hierarchical motion planning and control framework for overtaking maneuvers under racing circumstances. Unlike urban or highway autonomous driving conditions, race track driving requires longer prediction and planning horizons in order to respond to upcoming corners at high speed. In addition, the subject vehicle should determine the optimal action among possible driving modes when opponent vehicles are present. In order to meet these requirements and secure real time performance, a hierarchical architecture for decision making, motion planning, and control for an autonomous racing vehicle is proposed. The supervisor determines whether the subject vehicle should stay behind the preceding vehicle or overtake, and its direction when overtaking. Next, a high level trajectory planner generates the desired path and velocity profile in a receding horizon fashion.

This paper presents an integrated control of in-wheel motor (IWM) and electronic limited slip differential (eLSD) to enhance the vehicle lateral stability and maneuverability. The two actuators are utilized in the proposed controller to achieve separate purposes. The IWM controller is designed to modify the understeer gradient for enhanced handling characteristic and maneuverability. The eLSD controller is devised to improve the lateral stability to prevent oversteer in a severe maneuver. The proposed controller consists of a supervisor, upper-level controller and lower-level controller. The supervisor determines a target motion based on a target understeer gradient for IWM control and a yaw rate reference for eLSD control. The upper-level controller generates a desired yaw moment for the target motion. In the lower-level controller, the desired yaw moment is converted to the control inputs for IWMs in the two front wheels and eLSD at the rear axle.

This paper presents the research related to the self-driving system that has been actively carried out recently. Previous studies have been limited to ensure the path following performance in linear and steady state-alike handling region with small lateral acceleration. However, in the high speed driving, the vehicle cornering response is extended to nonlinear region where tire grips are saturated. This requires a technology to create the driving path for minimum time maneuvering while grasping the tire grip limits of the vehicle in real time. The entire controller consists of three stages-hierarchy: The target motion is determined in the supervisor phase, and the target force to follow the target behavior is calculated in the upper stage controller. Finally, the lower stage controller calculates the actuator phase control input corresponding to the target force.

A spark-ignition engine commonly induces tumble flow because it generates high turbulence, which is a crucial factor in determining the flame propagation speed. Since tumble affects not only the flame propagation speed but also the various in-cylinder phenomena, it predominantly determines the performance of the engine. In that sense, many studies have been conducted to investigate tumble. Although various studies have revealed the characteristics of tumble numerically and experimentally, there has been no research to identify the physical mechanisms of these characteristics. Although some studies specified the mechanisms from an angular momentum perspective, the theory was insufficient to explain the entire phenomena of tumble. Hence, this study attempts to comprehend the fundamental causes of tumble phenomena such as ‘spinning up’ and ‘vortex breakdown’ from the perspective of kinetic energy.

This paper describes a failsafe system of automated driving vehicles. The failsafe system consists of the following two parts: sliding mode observer-based environment sensor, chassis sensor fault detection, and emergency deceleration control. Two sliding mode observers are designed to reconstruct the fault of acceleration and environment sensor(Lidar) in a longitudinal direction. In the environment sensor's fault detection part, the longitudinal vehicle model receives clearance and relative velocity values. Therefore, failure diagnosis is possible regardless of environmental sensors, such as radar, lidar, and camera. This paper's sensor data is the failure of Delphi's Electronically Scanning Radar (ESR) and Ibeo's LUX Lidar installed in an autonomous vehicle. The emergency deceleration control algorithm employs the sliding mode control with adaptive convergence time. In the event of a failure, it is significant to control the vehicle within a short period safely.

This paper describes a lateral control of autonomous large size buses on road with large curvature. In the case of long and wide commercial vehicle such as large bus, applying centerline tracking controllers in constrained environments such as large curved road (e.g. turning at intersection) may cause some concerns. Two concerns are considered: inner lane crossing related to collisions with curb and opposite lane crossing related to threatening surrounding vehicles. Considering relations between width and curvature of the road and length and width of the large size bus, the curvature of road at which inner or outer lane crossing begin to occur was calculated when centerline tracking controller was applied. Thus, the proposed algorithm optimizes motion of the bus by using model predictive control (MPC) using road geometry as constraints.

Regulations regarding vehicles’ CO2 emissions are continuing to become stricter due to global warming. The CO2 regulations urge automobile manufacturers to develop gasoline engines with improved efficiency; however, the main obstacle to the improvement is the knock phenomenon in spark-ignition engines. If knock is predicted, the efficiency potential can be maximized in an engine by applying modest spark timing. Several research regarding knock prediction modeling have been conducted, and typically Livengood-Wu integral model is used to predict the knock occurrence. For the prediction, knock onset should be determined on a given pressure signal of given knock cycles for establishing the 0D ignition delay model. Several methodologies for knock onset determination have been developed because checking all the knock onset position by hand is impossible considering the breadth of data sets.

The recent development pace of the automotive technology is so rapid worldwide. Especially in a green car, hybrid electric vehicles (HEVs) have been studied a lot due to their significant effects on the urban driving. In the vehicle energy management strategy study, the driving speed is assumed to be known in advance, however the speed is not given in a real world. Accordingly, the prediction of vehicle speed is very important. In this study, we study the prediction methodology for the speed prediction using deep learning. Based on the vehicle driving speed data, the supervised deep learning has been used and the speed prediction accuracy using deep learning shows accurate results comparing to the actual speed. The supervised deep learning is used which is suitable for driving cycle database. As a result, the speed prediction after few seconds is feasible.

This paper proposes a rear-wheel steering control method that can modify and improve the vehicle lateral response without tire model and parameter. The proposed control algorithm is a combination of steady-state and transient control. The steady state control input is designed to modify steady-state yaw rate response of the vehicle, i.e. understeer gradient of the vehicle. The transient control input is a feedback control to improve the transient response when the vehicle lateral behavior builds up. The control algorithm has been investigated via computer simulations. Compared to classical control methods, the proposed algorithm shows good vehicle lateral response such as small overshoot and fast response. Specifically, the proposed algorithm can alleviate stair-shaped response of the lateral acceleration.

As emissions regulations become stricter, a variety of advanced combustion concepts that can reduce emissions with a higher thermal efficiency have been suggested. Dual-fuel combustion is one of the alternatives that has both premixed and non-premixed combustion characteristics. Knowing the effects of the mixture formation in dual-fuel combustion is important because it determines the ignition location and the following combustion phase. Hence, a thorough investigation on the related factors, such as the engine hardware or fuel spray, is required. Meanwhile, Computational Fluid Dynamics (CFD) is a good technique to visualize the in-cylinder phenomena and enables quantitative investigations into the detailed combustion characteristics. In this paper, a 3-dimensional CFD simulation was used to investigate the effects of the mixture formation in dual-fuel combustion. The combustion model consists of two parts.

In this study, the spalling issue in ball-type Constant Velocity Joints (CVJ) was investigated. As one of the most common types of outboard CVJ, a ball-type CVJ has spalling problems caused by fatigue at the internal contact points. It causes noise and vibration in vehicles, which results in CVJ failures. This study provides a spalling-estimation model for a ball-type CVJ, which was developed by the following five steps. First, the relative coordinates of the internal contact points between each component were established by forward kinematics. Second, the acting forces were calculated according to the results of the relative coordinate analyses and the vehicle driving conditions, and then normal pressure at the contact points was derived by Hertz contact theory. Third, the maximum sliding speeds at the contact points were also calculated using slip motion analyses. These normal pressure and maximum sliding speeds were used to estimate the shear stresses at the contact points.

The flame propagation characteristic is one of the greatest factor that determines the performance of spark ignition (SI) engines. The in-cylinder flow dynamics is very significant in terms of flame propagation because of its direct influence on the flame shape, turbulent flame speed, and the ignition quality. A number of different techniques are available to optimize the in-cylinder flow and maximize the utilization of turbulence for faster combustion, and tumble enhancement by intake port geometry is one of them. It requires excessive computational expenses to evaluate multiple designs under wide range of operating conditions by 3D-CFD, therefore, a low-dimensional model would be more competitive in such design optimization process. This work suggests a new modification approach for typical 0D turbulence model to take account for the tumble generation during the intake process as well as the turbulence characteristics associated with it.

Large-eddy simulation (LES) applications for internal combustion engine (ICE) flows are constantly growing due to the increase of computing resources and the availability of suitable CFD codes, methods and practices. The LES superior capability for modeling spatial and temporal evolution of turbulent flow structures with reference to RANS makes it a promising tool for describing, and possibly motivating, ICE cycle-to-cycle variability (CCV) and cycle-resolved events such as knock and misfire. Despite the growing interest towards LES in the academic community, applications to ICE flows are still limited. One of the reasons for such discrepancy is the uncertainty in the estimation of the LES computational cost. This in turn is mainly dependent on grid density, the CFD domain extent, the time step size and the overall number of cycles to be run. Grid density is directly linked to the possibility of reducing modeling assumptions for sub-grid scales.

Improving fuel efficiency and emission characteristics are significant issues in engine research. Because the engine has complex systems and various operating parameters, the experimental research is limited by cost and time. One-dimensional (1D) simulation has attracted the attention of researchers because of its effectiveness and relatively high accuracy. In a 1D simulation, the applied model must be accurate for the reliability of the simulation results. Because in-cylinder turbulence mainly determines the combustion characteristics, and mean flow velocity affects the in-cylinder heat transfer and efficiency in a spark-ignited (SI) engine, a number of sophisticated models have been developed to predict in-cylinder turbulence and mean flow velocity. In particular, tumble is a significant factor of in-cylinder turbulence in SI engine.

This paper aims to establish a systematic process of developing a brand driving sound. Firstly, principal factors of a brand sound identity are extracted from factor analysis of many sample cars. As a result, brand sound positioning map is drawn using jury test data. Also, the multiple regression analysis of subjective and objective test results is carried. As a result, the principal factors are expressed by objective test data and brand sound positioning map can be easily updated from the measurement data. In addition, what should be improved for designing a target sound is reviewed. Secondly, various technologies of target sound design are discussed to involve the brand identity and vehicle’s character in driving sound. Also, an efficient tool to implement the target sound with an active sound design (ASD) system in a vehicle is introduced. This tool enables to efficiently design, tune and simulate a target sound for ASD system in a laboratory.