Kwangsuck Boo

Model development and control methodology of a new electric power steering system

Abstract In this study, a new column-type electric power steering (EPS-TT) system is investigated. The remarkable features of this EPS-TT system are its opto-isolated torque sensor, which is used to make steering torque measurements, and its assist torque control methodology, which uses a unidirectional motor and two clutches. Thus it does not require a complicated motor drive system that consumes a large amount of electrical energy when the direction of rotation is reversed. This allows the new system to use a smaller and simpler assist motor. A full steering system model and a simplified model are developed to evaluate the EPS-TT system. A full car model is also used to investigate the vehicle response. A map-based control method and a proportional-integral-derivative control algorithm are designed to control the EPS-TT system. Various sinusoidal inputs are applied to the system and the resulting performance is analysed. The results show that the performance achieved by the EPS-TT system is similar to that of a conventional EPS system across the frequency domain. The results for the full steering system model are similar to those for the simplified model, but the vehicle response is slightly different. The map-based controller provided good performance without affecting the stability or controllability of the vehicle.

A study on an anti-lock braking system controller and rear-wheel controller to enhance vehicle lateral stability

Abstract This paper presents a mathematical vehicle model that is designed to analyse and improve the dynamic performance of a vehicle. A wheel slip controller for anti-lock braking system (ABS) brakes is formulated using a sliding mode controller and a proportional-integral-derivative (PID) controller for rear wheel steering is also designed to enhance the stability, steerability, and driveability of the vehicle during transient manoeuvres. The braking and steering performances of controllers are evaluated for various driving conditions, such as straight and J-turn manoeuvres. The simulation results show that the proposed full car model is sufficient to predict vehicle responses accurately. The developed ABS reduces the stopping distance and increases the longitudinal and lateral stability of both two-and four-wheel steering vehicles. The results also demonstrate that the use of a rear wheel controller as a yaw motion controller can increase its lateral stability and reduce the slip angle at high speeds.

Performance evaluation of traction control systems using a vehicle dynamic model

Abstract This study evaluates traction control systems (TCSs) composed of either a wheel slip controller or a throttle valve controller, or an integrated controller of both systems. The dynamic characteristics of a vehicle and a TCS are evaluated using a proposed full car model that can simulate the responses of both front-wheel-drive and four-wheel-drive vehicles. A driver model is also modified to control the vehicle during tests on a road with split Coefficients. The results show that the brake TCS provides more acceleration on uniform slippery and split roads, but the yaw rate and the lateral off set are larger than those obtained when an engine TCS is used. When the vehicle is cornering and accelerating with the brake or engine TCS, understeer or oversteer occur, depending on the driving conditions. An integrated TCS prevents most of these problems and improved the stability and controllability of the vehicle. Four-wheel-drive vehicles exhibit better traction control than two-wheel-drive vehicles, but their steerability is reduced.

Application of a sliding mode control to anti-lock brake system

This paper presents application method of a sliding mode wheel slip controller for ABS that improves the vehicle response and increases the safety on slippery road. Sliding mode wheel slip controller receives wheel slip ratio, vehicle velocity, longitudinal acceleration, and tire forces from vehicle model to generate braking pressures. In general, two solenoid valves, so called normal open and normal close valves, have been used to generate the hydraulic pressure necessary to generate proper braking force. In this paper, the valve operating method is proposed to implement the sliding mode control output in hydraulic unit of the ABS. The control performance is verified by using a HILS System that has a vehicle braking system controlled with On/Off solenoid valves.

A study on the welding seam tracking by using Laser Vision Sensor

This paper suggests one of the pre-measurement methods to trace a welding path. The welding path is measured by the LVS (Laser Vision Sensor). A line laser is projected on the weldment and a vision sensor captures images. Center points of the welding path are detected through image processing. Measured image coordinates are transformed to robot coordinates by coordinate transformations. The experimental structure is composed by 1-axis robot, cylinder, LVS. The rolling cylinder that is carved V-groove generates the continuous welding path and the 1-Axis robot follows the V-groove. The tracking algorithm is applied to generate the welding path that is obtained by the correlation of a set of groove center points. It is important that the robot traces the welding path accurately in the margin of error. Thus, the process of error data distinction and modification is implemented through the estimation algorithm.

Demonstration of non-collocated vibration control of a flexible manipulator using electrical dynamic absorbers

This paper describes an experimental study into the vibration control of a servo system comprising a servo motor and a flexible manipulator. Two modes of the system are controlled by using the servo motor and an accelerometer attached to the tip of the flexible manipulator. The control system is thus non-collocated. It consists of two electrical dynamic absorbers, each of which consists of a modal filter and, in case of an out-of-phase mode, a phase inverter. The experimental results show that each absorber acts as a mechanical dynamic vibration absorber attached to each mode and significantly reduces the settling time for the system response to a step input.

Nonlinear observer and robust controller design for enhancement of vehicle lateral stability

This paper describes the design of a sliding mode controller to control wheel slip. A yaw motion controller (YMC), which uses a PID control method, is also proposed for controlling the brake pressure of the rear and inner wheels to enhance lateral stability. It induces the yaw rate to track the reference yaw rate, and it reduces a slip angle on a slippery road. A nonlinear observer is also developed to estimate the vehicle variables difficult to measure directly. The braking and steering performances of the anti-lock brake system (ABS) and YMC are evaluated for various driving conditions, including straight, J-turn, and sinusoidal maneuvers. The simulation results show that developed ABS reduces the stopping distance and increases the longitudinal stability. The observer estimates velocity, slip angle, and yaw rate very well. The results also reveal that the YMC improves vehicle lateral stability and controllability when various steering inputs are applied. In addition, the YMC enhances the vehicle safety on a split-μ road.

Design of a current driver controller for MR CDC dampers

In this paper, a new current control algorithm has been proposed in order to control the current provided into a coil which produces a magnetic field in the MR (magneto-rheological) damper. The MR (magneto-rheological) damper is able to control a damping coefficient of the fluid through an orifice by using the MR fluid which has a characteristic of viscosity variation according to inducted current. In general, The MR CDC (continuous damping control) damper has faster response time than other ones. However, the MR damper has disadvantage of time delay according to input command due to very large inductance of internal coil. A new robust lead compensation control algorithm has been to overcome the above disadvantage of the MR damper. The proposed control algorithm is also able to provide more robust current control performance even under variation of the coil resistance induced from temperature variation.

Development of a vehicle height sensor for active suspension

This paper describes development method of a vehicle height sensor for active suspension controlled by using magnetic rheological fluid. Most of the current vehicle height sensors have been installed at external place of the damper and connected to that by linkage structures so far. This mechanism has disadvantage of instability and inaccuracy. The proposed sensor could be installed at inside of the MR CDC dampers so that there is no misalignment between the center axis of the damper and the vehicle height measuring position. The sensor mechanism has been designed based upon the well known magnetic field induced position sensor principle used in LVDT(linear variable displacement transducer). Dimensional parameters of the sensor could be determined by analyzing the magnetic phenomena around a sensor coil assembly and a sensor rod. Sensor output signal processing circuits have been also developed in order to have fast response and communicate with an active damper ECU. A series of experiments have been conducted to evaluate the steady state as well as the transient state performances. As the results, the developed sensors show that the accuracy is under plusmn0.5%, the resolution is 0.1 mm and the response time is up to 0.2 msec.

A study on real time integrated lane detection and vehicle tracking method with side-mirror cameras

Vehicle detection in real-time from rear-side of a host vehicle is one of important problems in Lane Change Assistance. In this paper, we propose a vision system for real-time vehicle and lane detection and tracking using two cameras, which are equipped under the wing mirror both left and right side. From the input images, EDLines algorithm is used for line segment detection in real-time. According to the achieved data, lane detection is developed by analyzing angles of the line segments, and area between two lanes on the same side of vehicle is defined. In the vehicle detection, based on the brightness and darkness between vehicle and road, vehicle is detected in real-time using the simple algorithm. Finally, kalman filter is used in vehicle tracking for vehicle information such as distance or speed.