A Longitudinal Model Based Probabilistic Fault Diagnosis Algorithm of Autonomous Vehicles Using Sliding Mode Observer

2017 ◽  
Author(s):  
Kwangseok Oh ◽  
Kyongsu Yi
Keyword(s):  
Fault Diagnosis ◽  
Autonomous Vehicles ◽  
Sliding Mode ◽  
Autonomous Vehicle ◽  
Sliding Mode Observer ◽  
Acceleration Sensor ◽  
Longitudinal Model ◽  
Longitudinal Acceleration ◽  
Relative Acceleration ◽  
Diagnosis Algorithm

This paper describes a longitudinal model based probabilistic fault diagnosis algorithm of autonomous vehicles using sliding mode observer. Autonomous vehicles use various sensors such as radar, lidar, and camera to obtain environment information. And internal sensors such as wheel speed, acceleration, and steering angle sensors have been used in vehicle to measure vehicle dynamic states. Based on the measured environment and vehicle states information, autonomous vehicle decides how to drive and control steering, throttle, and brake. Therefore, fault diagnosis of sensors used in autonomous vehicles is the most important for safe driving. In order to diagnosis longitudinal acceleration sensor fault of autonomous vehicle, longitudinal kinematic model has been used. The relative acceleration has been reconstructed using sliding mode observer based on environment information such as relative displacement and velocity between preceding vehicle and subject vehicle. The reconstructed relative acceleration has been used to compute longitudinal acceleration probabilistically based on analyzed longitudinal vehicle’s acceleration. The computed acceleration has been compared with measured acceleration for fault diagnosis of the acceleration sensor. The probabilistic fault diagnosis algorithm has been proposed and evaluated using actual data with arbitrary fault signal. The evaluation results of the proposed fault diagnosis algorithm show the reasonable fault diagnosis performance.

Fault Diagnosis of Exhaust Gas Recirculation and Variable Geometry Turbocharger Systems in a Passenger Car Diesel Engine Based on a Sliding Mode Observer for Air System States Estimation

2014 ◽  
Vol 136 (3) ◽  
Author(s):  
Hyunjun Lee ◽  
Joonhee Lee ◽  
Myoungho Sunwoo
Keyword(s):  
Fault Diagnosis ◽  
Sliding Mode ◽  
Sliding Mode Observer ◽  
Generation Model ◽  
Exhaust Gas ◽  
Variable Geometry ◽  
Variable Geometry Turbocharger ◽  
Diagnosis Algorithm ◽  
Gas Recirculation ◽  
Residual Generation

In this paper, we propose a sliding mode observer based fault diagnosis algorithm for diesel engines with exhaust gas recirculation (EGR) and variable geometry turbocharger (VGT) systems. The nonlinear sliding mode observer is proposed for precise states estimation of air system in diesel engines. Based on the estimation results of the observer and the limited sensor information in mass-produced engines, a residual generation model is derived. A modified cumulative summation algorithm is applied to the residual generation model for robust fault detection and isolation of the EGR and VGT systems. The proposed observer based fault diagnosis algorithm is implemented on a real-time embedded system, and the bypass function of an engine management system (EMS) is applied to generate multiple types of fault conditions in the systems. As a result of this study, estimation performance of the proposed observer is validated and successful fault diagnosis of the EGR and VGT systems is demonstrated through engine experiments.

A Predictive Approach to the Fault Detection in Fail-Safe System of Autonomous Vehicle Based on the Multi-Sliding Mode Observer

2018 ◽  
Author(s):  
Kwangseok Oh ◽  
Sungyoul Park ◽  
Kyongsu Yi
Keyword(s):  
Fault Detection ◽  
Autonomous Vehicles ◽  
Sliding Mode ◽  
Kinematic Model ◽  
Detection Algorithm ◽  
Sliding Mode Observer ◽  
Longitudinal Acceleration ◽  
The Subject ◽  
Fail Safe ◽  
Relative States

This paper describes a predictive method for fault detection in the fail-safe system of autonomous vehicles based on the multi sliding mode observer. In order to detect faults in sensors, such as radar and acceleration sensors used for longitudinal control of the autonomous vehicles, the kinematic model-based sliding mode observer and a predictive algorithm have been used. The driving condition that the subject vehicle is driving with a preceding vehicle has been considered in this study. The relative acceleration has been reconstructed based on the sliding mode observer using relative displacement and velocity. Based on the reconstructed relative acceleration, the upper and lower limits of longitudinal acceleration for fault detection have been derived based on the stochastic analysis of the driver’s driving data. The measured longitudinal acceleration of the subject vehicle has been used to predict the relative states using the longitudinal kinematic model. The predicted relative states have been stored, and the stored states that represent the current states have been used to detect faults in the sensors. With regard to longitudinal acceleration, the multi sliding mode observer has been used to detect faults in the acceleration sensor. The predictive fault detection algorithm proposed in this study can detect faults in the environment sensors individually based on past sensor information. In order to obtain a reasonable performance evaluation, actual driving data and a 3D full vehicle model constructed in the Matlab/Simulink environment have been used in this study. The results of the performance evaluation show that the predictive fault detection algorithm was successfully able to detect faults in the sensors for longitudinal control individually.

Robust actuator and sensor fault reconstruction of wind turbine using modified sliding mode observer

2018 ◽  
Vol 41 (6) ◽  
pp. 1504-1518 ◽  
Author(s):  
Mostafa Rahnavard ◽  
Moosa Ayati ◽  
Mohammad Reza Hairi Yazdi
Keyword(s):  
Fault Diagnosis ◽  
Wind Turbine ◽  
Rotational Speed ◽  
Sliding Mode ◽  
Sliding Mode Observer ◽  
Sensor Faults ◽  
Actuator Faults ◽  
Fault Reconstruction ◽  
Different Order ◽  
Aerodynamic Torque

This paper proposes a robust fault diagnosis scheme based on modified sliding mode observer, which reconstructs wind turbine hydraulic pitch actuator faults as well as simultaneous sensor faults. The wind turbine under consideration is a 4.8 MW benchmark model developed by Aalborg University and kk-electronic a/s. Rotor rotational speed, generator rotational speed, blade pitch angle and generator torque have different order of magnitudes. Since the dedicated sensors experience faults with quite different values, simultaneous fault reconstruction of these sensors is a challenging task. To address this challenge, some modifications are applied to the classic sliding mode observer to realize simultaneous fault estimation. The modifications are mainly suggested to the discontinuous injection switching term as the nonlinear part of observer. The proposed fault diagnosis scheme does not require know the exact value of nonlinear aerodynamic torque and is robust to disturbance/modelling uncertainties. The aerodynamic torque mapping, represented as a two-dimensional look up table in the benchmark model, is estimated by an analytical expression. The pitch actuator low pressure faults are identified using some fault indicators. By filtering the outputs and defining an augmented state vector, the sensor faults are converted to actuator faults. Several fault scenarios, including the pitch actuator low pressure faults and simultaneous sensor faults, are simulated in the wind turbine benchmark in the presence of measurement noises. Simulation results show that the modified observer immediately and faithfully estimates the actuator faults as well as simultaneous sensor faults with different order of magnitudes.

scholarly journals Estimation and Fault Diagnosis of Lithium-Ion Batteries: A Fractional-Order System Approach

2018 ◽  
Vol 2018 ◽  
pp. 1-12 ◽  
Author(s):  
Shulan Kong ◽  
Mehrdad Saif ◽  
Guozeng Cui
Keyword(s):  
Fault Diagnosis ◽  
Fault Detection ◽  
Fractional Order ◽  
Sliding Mode ◽  
Estimation Error ◽  
Lithium Ion ◽  
Fault Isolation ◽  
Sliding Mode Observer ◽  
Fault Detection And Isolation ◽  
Order System

This study investigates estimation and fault diagnosis of fractional-order Lithium-ion battery system. Two simple and common types of observers are designed to address the design of fault diagnosis and estimation for the fractional-order systems. Fractional-order Luenberger observers are employed to generate residuals which are then used to investigate the feasibility of model based fault detection and isolation. Once a fault is detected and isolated, a fractional-order sliding mode observer is constructed to provide an estimate of the isolated fault. The paper presents some theoretical results for designing stable observers and fault estimators. In particular, the notion of stability in the sense of Mittag-Leffler is first introduced to discuss the state estimation error dynamics. Overall, the design of the Luenberger observer as well as the sliding mode observer can accomplish fault detection, fault isolation, and estimation. The effectiveness of the proposed strategy on a three-cell battery string system is demonstrated.

scholarly journals Rolling-Element Bearing Fault Diagnosis Using Advanced Machine Learning-Based Observer

2019 ◽  
Vol 9 (24) ◽  
pp. 5404 ◽  
Author(s):  
Farzin Piltan ◽  
Alexander E. Prosvirin ◽  
Inkyu Jeong ◽  
Kichang Im ◽  
Jong-Myon Kim
Keyword(s):  
Machine Learning ◽  
Fault Diagnosis ◽  
Sliding Mode ◽  
Sliding Mode Observer ◽  
Rolling Element Bearing ◽  
Fault Estimation ◽  
Detection And Identification ◽  
Rolling Element ◽  
Fuzzy Sliding Mode ◽  
Element Bearing

Rotating machines represent a class of nonlinear, uncertain, and multiple-degrees-of-freedom systems that are used in various applications. The complexity of the system’s dynamic behavior and uncertainty result in substantial challenges for fault estimation, detection, and identification in rotating machines. To address the aforementioned challenges, this paper proposes a novel technique for fault diagnosis of a rolling-element bearing (REB), founded on a machine-learning-based advanced fuzzy sliding mode observer. First, an ARX-Laguerre algorithm is presented to model the bearing in the presence of noise and uncertainty. In addition, a fuzzy algorithm is applied to the ARX-Laguerre technique to increase the system’s modeling accuracy. Next, the conventional sliding mode observer is applied to resolve the problems of fault estimation in a complex system with a high degree of uncertainty, such as rotating machinery. To address the problem of chattering that is inherent in the conventional sliding mode observer, the higher-order super-twisting (advanced) technique is introduced in this study. In addition, the fuzzy method is applied to the advanced sliding mode observer to improve the accuracy of fault estimation in uncertain conditions. As a result, the advanced fuzzy sliding mode observer adaptively improves the reliability, robustness, and estimation accuracy of rolling-element bearing fault estimation. Then, the residual signal delivered by the proposed methodology is split in the windows and each window is characterized by a numerical parameter. Finally, a machine learning technique, called a decision tree, adaptively derives the threshold values that are used for problems of fault detection and fault identification in this study. The effectiveness of the proposed algorithm is validated using a publicly available vibration dataset of Case Western Reverse University. The experimental results show that the machine learning-based advanced fuzzy sliding mode observation methodology significantly improves the reliability and accuracy of the fault estimation, detection, and identification of rolling element bearing faults under variable crack sizes and load conditions.

Sign in / Sign up

Export Citation Format

Share Document