Search

This study presents sensitivity-based adaptive model-free adaptive displacement and velocity control algorithms for unknown single-input multi-output (SIMO) systems using recursive least squares (RLS) with exponential forgetting. Nonlinearities in real-world systems make predictions of the dynamic behaviors of systems challenging and have a negative influence on the performance of control systems. To address this problem, the sensitivity-based model-free adaptive control algorithm, which does not require a mathematical model of a system, is proposed in this study. In the algorithm, a feedback control law for a SIMO system was designed with multiple adaptive feedback gains. The virtual function that represents the relationship between multiple feedback gains and control errors was also designed to construct an adaptation rule with a linear combination. The coefficients in the virtual function were estimated in real-time using RLS with multiple exponential forgetting factors. By using the estimated coefficients, the adaptation rule for feedback gains with the gradient descent method was designed, which can be automatically adjusted through applying Lyapunov’s direct method. The performance of the proposed model-free adaptive control algorithm was evaluated in an actual DC motor system under angular displacement and velocity tracking control scenarios. Evaluation results show that the designed control algorithm enables the system to track the desired output successfully without system modeling.

In this paper, a self-tuning rule-based position control algorithm is proposed for DC motors with system parameter estimation using the recursive least squares method. First, a mathematical model of the angular position control of a DC motor was derived. Next, the time-varying parameters including the rotational inertia in the model were estimated using the RLS method along with multiple forgetting factors without prior knowledge of the system. Based on the derived model and the parameter estimation, a sliding mode control algorithm was designed by applying a self-tuning rule that enables the magnitude of the voltage input to be adaptively adjusted for improvement of the energy efficiency. The performance of the designed control algorithm was then experimentally evaluated under several different load conditions. Finally, the evaluation results show that the designed controller achieves a satisfactory capability for a DC motor to deal with both tracking accuracy and energy efficiency without prior knowledge of the system.

This paper describes model-based separated fault detection and fault tolerant control of longitudinal autonomous driving using dual-sliding mode observer for functional safety. Internal and environment sensors such as camera or radar are required to measure the acceleration information of the subject vehicle and the relative distance and velocity information between the preceding and subject vehicles in longitudinal autonomous driving. In order to detect the independent fault of each sensor, a dual-sliding mode observer (SMO) is used for fault reconstruction under the assumption that V2V (Vehicle to Vehicle) communication for vehicle driving state is available. The each SMO reconstructs the expected fault in sensor based on discontinuous injection term used for converging output error to zero. Based on the reconstructed fault by each SMO, faults are detected using threshold approach. When the fault is detected, the reconstructed fault is used for fault tolerant control by subtracting to faulty data. The proposed fault detection (FD) and fault tolerant control (FTC) algorithms were evaluated using actual driving data and a three-dimensional (3D) vehicle model with a linear quadratic regulator for following control. The evaluation results are presented and analyzed with regard to fault reconstruction, detection, and tolerant control in four cases wherein two types of faults were applied.

This paper presents a novel design and control method of an active suspension system using a linear pump control–based hydraulic system for a cost-effective application. Various active suspension systems have been proposed and applied to vehicles due to its ability to improve ride comfort and handling performance even though these active suspension systems are not popular because of their complexity, high cost, heavyweight, and low power efficiency. A new type of active suspension actuator system was designed and validated herein based on the methods of actuator sizing and modified control scheme to address the aforementioned issues. System power capability has been analyzed under various dynamics and road conditions. Active suspension actuator components have been designed based on the results. The electro-hydraulic system is powered by a battery to reduce the energy consumption of the system; hence, it is operated by torque on demand. A double-acting linear hydraulic motor pump with a dual rack and pinion has been proposed for hydraulic force control with a simple on/off switch operation. The actuator force has been controlled by continuous linear motor pump displacement control via torque control using a three-phase synchronous brushless alternative current motor. Dynamic performance evaluation of the actuator system has been conducted using AMESIM and actual rig test. Active height and roll control algorithms for the enhancement of vehicle dynamics considering actuator dynamics have also been developed and validated through the rig and real vehicle tests. The evaluation results showed that the linear motor pump–based active suspension system performs as well as the previous complicated hydraulic active suspension system. The new active system proposed in this study was able to improve the vehicle dynamics using cost-effective actuation system significantly.

This paper presents a model predictive control-based fault detection and reconstruction algorithm for longitudinal control of autonomous driving using a multi-sliding mode observer. In order to secure the safe longitudinal control of a vehicle, a numbers of factors must be ensured, such as the reliability of the longitudinal information, the data on the forward object from the environment sensor, and the acceleration of the ego vehicle. Thus, we propose a reasonable failure detection scheme for the acceleration signal of the host vehicle and the relative values of the front object of the radar. In order to identify the faults of the radar and the vehicle acceleration sensor related to the automated longitudinal control, the multiple sliding mode observer and prediction of model predictive control (MPC) algorithm are applied. The relative acceleration is reconstructed by applying a sliding mode observer (SMO) with clearance and relative speed measurements. The upper and lower limits of longitudinal acceleration were computed by analyzing human driving data under the preceding vehicle and reconstructed acceleration. A proper acceleration range can be defined precisely based on several reconstructed upper and lower bounds by using a multiple sliding mode observer with stored prediction data of relative values, making it possible to effectively identify the fault of the host vehicle’s acceleration sensor. By applying MPC for this study, optimal control input and prediction of relative states can be obtained that are more reasonable than those using the linear prediction model. The proposed fault detection algorithm can identify the abnormal state of the environment sensors by using the accumulated past sensor data. By comparing the stored prediction of relative states with the stored data on current states for a given period, the signal faults of the longitudinal target information can be detected from environment sensors. With these fault indices of states, the final fault diagnoses of sensors can be determined by assessing confidence through statistical analysis of 27 sets of normal driving data. In order to obtain a reasonable performance evaluation, this study uses actual driving data and a 3D full vehicle model constructed in the MATLAB/Simulink environment. The test results reveal that the proposed algorithm can successfully detect the fault of the radar and acceleration sensor of the automated driving vehicle.

Lateral thrust jets exhibit better maneuverability performance than control surfaces such as conventional fins for attitude control or the orbital transfer of guided weapons. In the supersonic region, however, jet interaction phenomena occur due to the lateral thrust jet during flight and a complicated flow structure is exhibited by the interaction of the shock wave, boundary layer flow, and vortex flow. In particular, hit-to-kill interceptors require precise control and maneuvering; therefore, it is necessary to analyze the effect of the jet interaction flow. A number of conventional jet interaction analyses have been performed under low-altitude conditions; however, there are not many cases with respect to medium-altitude conditions. Unlike low-altitude conditions, jet interaction flows at medium-altitude conditions have different flow characteristics. In this study, a jet interaction flow analysis is performed on a lateral thrust jet controlled interceptor operating at medium altitudes. Based on the results, the characteristics of the flow structure and the changes in the aerodynamic coefficients are analyzed.

This paper describes a yaw stability control algorithm of 4WD vehicles based on model predictive torque vectoring with physical constraints. A vehicle planar model based predictive rear and all-wheel torque vectoring algorithms were developed for 4WD vehicles by considering predictive states and driver’s steering wheel angle. The physical constraints applied to the model predictive control consist of three types: limitation on magnitude of tire force, change rate of tire force, and output torque of transfer case. Two types of torque vectoring algorithms, rear-wheel and all-wheel, were constructed for comparative analysis. The steady state yaw rate was derived and applied as a desired value for yaw stability of the vehicle. The algorithm was constructed in a MATLAB/Simulink environment and the performance evaluation was conducted under various test scenarios, such as step steering and double lane change, using the CarSim software. The evaluation results of the predictive torque vectoring showed sound performance based on the prediction of states and driver’s steering angle.

This paper describes state estimation and tracking control algorithms for use in an autonomous truck using a single-wheel driving module and simulation-based performance evaluation with actual data from a one-layer laser scanner. In order to construct the tracking control algorithm, the one-layer laser scanner has been used to detect the object as two-dimensional point cloud data. The tracking control algorithm consists of three routines—perception, decision, and control. In the perception routine, the coordinate transformation, downsizing, clustering, and state estimation have been conducted for calculation of the object states, such as position and velocity, using point data from the one-layer laser scanner. The velocity components of the object have been estimated based on the extended Kalman filter, and the desired straight path has been derived using the estimated velocity of the object. To track the object reasonably, lateral error and yaw angle error have been defined with respect to the desired straight path. The error dynamics has been derived using a single-wheel driving module, equipped with a planar truck model based on the vehicle dynamics. Using the derived planar truck model, the optimal steering input for tracking the object has been computed based on the linear quadratic regulator. The actual point cloud data from the one-layer laser scanner has been used to conduct reasonable performance evaluation of the tracking control algorithm proposed in this study. The actual data-based performance evaluation of the algorithm has been conducted in the MATLAB/SIMULINK environment with various scenarios. The performance evaluation results show that the tracking control algorithm proposed in this study can manipulate the single-wheel driving module of the autonomous truck to track the object soundly, based on the state estimation algorithm.

This study focuses on the development and evaluation of advanced safety algorithms for excavators using virtual reality (VR). An excavator typically operates under a stationary state with its working parts rotating 360° in coordination with nearby workers. During excavation, a fatal accident can occur due to operator carelessness and work site blind spots. Accidents due to collisions with nearby workers have been increasing. Accordingly, we presented safety system algorithms to prevent collisions with surrounding objects and secure the maximum working area in this study. We also evaluated the performance of safety system algorithms using VR. For risk assessment, we calculated the predicted working area through a kinematic analysis of the excavator’s working parts and accordingly conducted target selection of risk factors. We used time-to-collision and warning indices as safety indices for the safety assessment of the selected target and divided the excavator’s working modes into three categories: Safe, warning, and emergency braking modes. Control inputs, such as alarms and braking, were appropriately defined for each working mode. Under warning mode, workers can avoid collisions because a safety system will alert them of dangerous situations through an alarm. Under emergency braking mode, an emergency braking input signal is dispatched with an alarm to automatically prevent collisions. The advanced safety algorithm proposed in the study was developed in MATLAB/Simulink environment. The VR application was developed using a physics engine. On the basis of this application, the performance evaluation of the safety system algorithms was conducted on the frequently occurring sticking scenario.

This study proposes algorithms for performance index-based actuator fault detection and fault-tolerant control with adaptive feedback. Longitudinal control algorithm of autonomous vehicles is generally divided into three main parts such as supervisory, upper level, and lower level controllers. In the supervisory controller part, desired behavior is determined based on control targets. Therefore, desired vehicle motion such as longitudinal acceleration can be designed in the upper level controller part using various control methods. In the lower level controller part, actuator control input is determined for tracking the desired vehicle motion designed in the upper level controller part. In order for fault detection and fault-tolerant control of actuators used for longitudinal autonomous driving, adaptive feedback controller has been designed with the MIT rule for determination of the desired longitudinal acceleration and the control input has been compared with the actual acceleration for performance index-based fault detection. It is designed that the adaptive feedback controller adjust the desired acceleration for making reasonable desired behavior despite of existence of actuator fault or performance degradation. The window-based weighted standard deviation of error between the adaptive desired longitudinal acceleration and current acceleration of vehicle has been used for computation of performance index. Performance evaluation has been conducted using Matlab/Simulink and commercial software (CarMaker).

The study deals with the design of control algorithms to enhance the steering performance of all-terrain cranes. For this, its hydraulic steering system was modeled using a multiphysics simulation tool and the self-tuning fuzzy PID controller was developed to control hydraulic steering actuators with nonlinearities. The designed control algorithm was validated through co-simulation using a real-time simulator and showed satisfactory performance.

This paper proposes sensing and safety techniques for autonomous excavators working on construction sites. In autonomous construction equipment, safety is a significant concern to reduce fatal accidents and machinery damage. In taking up this point of view, this paper deals with object tracking, motion prediction, and track management for detected objects that can improve the safety function of autonomous construction equipment. By using the info of predicted motion of detected objects and actual working areas of an excavator, and calculating safety indices, the proposed safety algorithms can evaluate the degree of safety (i.e., potential collision risks).

This paper proposes an adaptive prediction time horizon based model predictive steering control algorithm for autonomous vehicle with disturbance estimation. The model predictive control (MPC) requires relatively accurate system model to compute the reasonable control inputs. There exists model uncertainty between system model and actual system. And the uncertainty can be increased in the prediction step of MPC and that can have an influence on control performance. In order to address the issue, the uncertainty has been estimated using the sliding mode observer and predicted in the designed prediction algorithm based on gray prediction model. Using the predicted future uncertainties, the number of prediction steps has been determined with the uncertainty threshold value. The path tracking control algorithm for autonomous driving has been used to compute the steering angle of vehicle. The performance evaluation of the adaptive MPC has been conducted using the simplified bicycle model in Matlab/Simulink environment under the lane change scenario.

This paper presents an adaptive sliding mode observer for input fault reconstruction of longitudinal autonomous driving. Sliding mode observer is the robust observer against disturbance, which is used to reconstruct the fault and state estimation. In order to design the injection parameter for sliding mode observer, the boundary of errors that include the fault is required. However, it is difficult to expect the fault magnitude for design the injection parameter. The proposed method is to estimate the proportional constant from the relationship between output error and injection parameter based on recursive least squares. Then, it is used to update the adaptive parameter based on MIT rule. The performance evaluation algorithm was conducted in Matlab/Simulink environment using actual longitudinal driving data and 3-dimensions vehicle model with the applied various faults.

This paper describes derivation algorithm and evaluation results of a Poincare-Bendixson theorem based target acceleration computation algorithm for autonomous driving on inverse time to collision and time headway plane. Derivation of target acceleration is needed for longitudinal autonomous driving. Ellipsoidal driving area is derived for considering driver’s driving characteristic and safety in time headway-inverse time to collision (TTC) plane. And target acceleration computation algorithm has been proposed based on Poincare-Bendixson theorem. Ellipsoidal driving areas are divided main driving area and real-time driving area. Main driving area is derived based on limit of inverse TTC and time headway for takeover time and human factor, real-time driving area is derived through current driving point with ratio of main driving area. It is designed to computation the target acceleration after deriving the target direction by applying a specific angle based on the normal to the current driving point through the real-time driving area. Specific angle is arbitrary value applied acceleration limitation of actual vehicle. The performance evaluation of target acceleration computation algorithm is has been conducted in Matlab/Simulink environment. It is expected that the proposed algorithm can be used for longitudinal control algorithm for safety and personalization of autonomous vehicle.

This paper investigates on sensor fault reconstruction of sensors used for steering control of autonomous vehicle for functional safety. Sensor information such as steering angle and longitudinal velocity is generally needed for the design of steering feedback control system. If there exists unexpected fault signals in sensors, fatal accident can occur during autonomous driving because controller cannot compute the accurate control input. In this study, the sliding mode observer has been designed for fault reconstruction of steering angle and velocity sensors. In order to design the observer, the bicycle model that represents dynamic relationship between steering angle and velocities such as lateral velocity and yaw rate of vehicle has been used. The stability analysis has been conducted in accordance with velocity of the vehicle. The fault signals in sensors have been reconstructed using the injection term in sliding mode observer with the sliding mode gains designed for the stability. The performance evaluation has been conducted in Matlab/Simulink environment under the curved path tracking scenario.

The capability and diversity of high-performance microprocessors is increasing with each process technology generation to meet increasing application demands. The cooling designs for these chips must deal with larger temperature gradients across the die than previous generations. Dissipation of the thermal energy from heat generating parts to a heat sink via conduction occurs through a thermal interfacial material (TIM). One of the most important parameters of TIMs is their thermal conductivity that impacts the efficiency of the thermal management. Therefore, in this study, three new different types of TIMs with high thermal conductivity (> 20W/mK) were examined.In this paper, workability and reliability test results will be shown for metal sheet TIM (60W/mK), graphite sheet TIM (24W/mK) and silver (Ag) sintered TIM (50W/mK) materials. All test package sizes are covered with 52.5-mm body size or 45-mm body size for large package with > 45-mm body size in a flip chip ball grid array (FCBGA) package. The reliability test items are high-temperature storage (HTS) @ 150’C 1000hrs, temperature cycling (TC) ‘B’ 1000X and unbiased highly accelerated stress test (uHAST) for 96hrs. Open/short (O/S) test, scanning acoustic tomography (SAT) or X-ray analysis data are reported. Post reliability test, lid pull test results will be shown as well.

본 연구는 슬라이딩 모드 관측 및 제어 기반 자율주행 자동차의 종방향 고장 허용 제어 알고리즘에 관한 것이다. 종방향 가속도 센서의 고장 탐지를 위해 슬라이딩 모드 관측기를 이용하여 상대가속도를 재건한다. 재건된 상대가속도 기반 가속도 확률분포를 이용하여 종방향 가속도의 확률적 고장탐지를 수행한다. 고장 탐지 시 재건된 가속도를 이용하여 가상 가속도를 도출하고, 가상 가속도를 허용 제어를 위해 하위 제어기의 가속도 입력으로 사용한다. 상위 제어기의 슬라이딩 모드 제어기는 선행차량 가속도, 가상 가속도, 하위 제어기 제어성능 불확실성에 대해 강건한 요구 가속도를 도출한다. 성능평가를 위해 속도 구간 별 실 주행 데이터와 3차원 차량 동역학 모델을 이용하여 사용하였다. 본 연구에서 제안하는 고장 허용 제어 알고리즘은 향후 자율주행 종방향 고장안전을 위해 적용될 것으로 예상한다.

This paper describes a functional perspective-based probabilistic fault detection and diagnostic algorithm of an autonomous vehicle using a longitudinal kinematic model. The relative displacement and velocity between the subject vehicle and a preceding vehicle was obtained by a radar installed in front of the autonomous vehicle. The obtained relative values were used to control the longitudinal behavior of the autonomous vehicle, the longitudinal acceleration of which was obtained from an internal sensor. In order to detect and diagnose actual faults in the obtained values, such as relative displacement, velocity, and acceleration, a fault detection and diagnostic algorithm for the longitudinal control of the autonomous vehicle was developed using a sliding mode observer and predictive function. The probabilistic analysis of fault signals was conducted using the constructed sliding mode observer and predictive function. The actual driving data of the vehicle preceding the subject vehicle was used for the rational performance evaluation of the proposed algorithm. The performance evaluation was conducted in the MATLAB/SIMULINK environment. The evaluation results showed that the proposed fault detection and diagnostic algorithm can stochastically detect and diagnose applied fault signals.

This paper presents a laser scanner–based predictive safety system for excavators. Blind spots on excavators and operators’ carelessness cause majority of fatal accidents, such as those in which de...

This research deals with assist wheel control to improve the steering efficiency and dynamic stability of the multi-axle crane based on model predictive control. Since multi-axle crane has relatively high inertia and long distance between the axles, it has slow dynamic response, and thus, different steering strategies according to driving speed intervals are required. Specifically, the steering strategy is that the number of wheels fixed mechanically increases to secure dynamic stability as the driving speed increases. However, although this strategy enables to secure stability by slowing down the dynamic response, it also has a weakness to decrease the steering efficiency. If the steering efficiency is decreased, it may result in augmenting an accident rate due to an increase in driver’s fatigue. Therefore, this study suggests a new steering strategy to improve the steering efficiency by simultaneously guaranteeing dynamic stability. The suggested steering control algorithm can enhance both steering efficiency and stability by deriving an assist wheel control input based on model predictive control using a variable weighting factor derived from the driver’s steering input (steering angle on the first axle). Development and performance evaluation of the suggested algorithm were conducted in the MATLAB/Simulink environment, and evaluation results confirmed that the developed algorithm could both enhance the driver’s steering efficiency and secure dynamic stability.